From db83705db748f2c207816f4eccce429c98823889 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:48:14 +0000 Subject: [PATCH 01/31] =?UTF-8?q?=E6=96=B0=E5=BB=BA=20torchreid.egg-info?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep | 0 1 file changed, 0 insertions(+), 0 deletions(-) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep new file mode 100644 index 0000000000..e69de29bb2 -- Gitee From e942a07cce7d5fad43aa46ad8725492ea59d20cb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:48:44 +0000 Subject: [PATCH 02/31] init --- .../torchreid.egg-info/PKG-INFO | 330 ++++++++++++++++++ .../torchreid.egg-info/SOURCES.txt | 87 +++++ .../torchreid.egg-info/dependency_links.txt | 1 + .../torchreid.egg-info/requires.txt | 16 + .../torchreid.egg-info/top_level.txt | 1 + 5 files changed, 435 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO new file mode 100644 index 0000000000..ea44b1cca6 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO @@ -0,0 +1,330 @@ +Metadata-Version: 2.1 +Name: torchreid +Version: 1.4.0 +Summary: A library for deep learning person re-ID in PyTorch +Home-page: https://github.com/KaiyangZhou/deep-person-reid +Author: Kaiyang Zhou +License: MIT +Keywords: Person Re-Identification,Deep Learning,Computer Vision +Platform: UNKNOWN +License-File: LICENSE + +Torchreid +=========== +Torchreid is a library for deep-learning person re-identification, written in `PyTorch `_ and developed for our ICCV'19 project, `Omni-Scale Feature Learning for Person Re-Identification `_. + +It features: + +- multi-GPU training +- support both image- and video-reid +- end-to-end training and evaluation +- incredibly easy preparation of reid datasets +- multi-dataset training +- cross-dataset evaluation +- standard protocol used by most research papers +- highly extensible (easy to add models, datasets, training methods, etc.) +- implementations of state-of-the-art deep reid models +- access to pretrained reid models +- advanced training techniques +- visualization tools (tensorboard, ranks, etc.) + + +Code: https://github.com/KaiyangZhou/deep-person-reid. + +Documentation: https://kaiyangzhou.github.io/deep-person-reid/. + +How-to instructions: https://kaiyangzhou.github.io/deep-person-reid/user_guide. + +Model zoo: https://kaiyangzhou.github.io/deep-person-reid/MODEL_ZOO. + +Tech report: https://arxiv.org/abs/1910.10093. + +You can find some research projects that are built on top of Torchreid `here `_. + + +What's new +------------ +- [Aug 2021] We have released the ImageNet-pretrained models of ``osnet_ain_x0_75``, ``osnet_ain_x0_5`` and ``osnet_ain_x0_25``. The pretraining setup follows `pycls `_. +- [Apr 2021] We have updated the appendix in the `TPAMI version of OSNet `_ to include results in the multi-source domain generalization setting. The trained models can be found in the `Model Zoo `_. +- [Apr 2021] We have added a script to automate the process of calculating average results over multiple splits. For more details please see ``tools/parse_test_res.py``. +- [Apr 2021] ``v1.4.0``: We added the person search dataset, `CUHK-SYSU `_. Please see the `documentation `_ regarding how to download the dataset (it contains cropped person images). +- [Apr 2021] All models in the model zoo have been moved to google drive. Please raise an issue if any model's performance is inconsistent with the numbers shown in the model zoo page (could be caused by wrong links). +- [Mar 2021] `OSNet `_ will appear in the TPAMI journal! Compared with the conference version, which focuses on discriminative feature learning using the omni-scale building block, this journal extension further considers generalizable feature learning by integrating `instance normalization layers `_ with the OSNet architecture. We hope this journal paper can motivate more future work to taclke the generalization issue in cross-dataset re-ID. +- [Mar 2021] Generalization across domains (datasets) in person re-ID is crucial in real-world applications, which is closely related to the topic of *domain generalization*. Interested in learning how the field of domain generalization has developed over the last decade? Check our recent survey in this topic at https://arxiv.org/abs/2103.02503, with coverage on the history, datasets, related problems, methodologies, potential directions, and so on (*methods designed for generalizable re-ID are also covered*!). +- [Feb 2021] ``v1.3.6`` Added `University-1652 `_, a new dataset for multi-view multi-source geo-localization (credit to `Zhedong Zheng `_). +- [Feb 2021] ``v1.3.5``: Now the `cython code `_ works on Windows (credit to `lablabla `_). +- [Jan 2021] Our recent work, `MixStyle `_ (mixing instance-level feature statistics of samples of different domains for improving domain generalization), has been accepted to ICLR'21. The code has been released at https://github.com/KaiyangZhou/mixstyle-release where the person re-ID part is based on Torchreid. +- [Jan 2021] A new evaluation metric called `mean Inverse Negative Penalty (mINP)` for person re-ID has been introduced in `Deep Learning for Person Re-identification: A Survey and Outlook (TPAMI 2021) `_. Their code can be accessed at ``_. +- [Aug 2020] ``v1.3.3``: Fixed bug in ``visrank`` (caused by not unpacking ``dsetid``). +- [Aug 2020] ``v1.3.2``: Added ``_junk_pids`` to ``grid`` and ``prid``. This avoids using mislabeled gallery images for training when setting ``combineall=True``. +- [Aug 2020] ``v1.3.0``: (1) Added ``dsetid`` to the existing 3-tuple data source, resulting in ``(impath, pid, camid, dsetid)``. This variable denotes the dataset ID and is useful when combining multiple datasets for training (as a dataset indicator). E.g., when combining ``market1501`` and ``cuhk03``, the former will be assigned ``dsetid=0`` while the latter will be assigned ``dsetid=1``. (2) Added ``RandomDatasetSampler``. Analogous to ``RandomDomainSampler``, ``RandomDatasetSampler`` samples a certain number of images (``batch_size // num_datasets``) from each of specified datasets (the amount is determined by ``num_datasets``). +- [Aug 2020] ``v1.2.6``: Added ``RandomDomainSampler`` (it samples ``num_cams`` cameras each with ``batch_size // num_cams`` images to form a mini-batch). +- [Jun 2020] ``v1.2.5``: (1) Dataloader's output from ``__getitem__`` has been changed from ``list`` to ``dict``. Previously, an element, e.g. image tensor, was fetched with ``imgs=data[0]``. Now it should be obtained by ``imgs=data['img']``. See this `commit `_ for detailed changes. (2) Added ``k_tfm`` as an option to image data loader, which allows data augmentation to be applied ``k_tfm`` times *independently* to an image. If ``k_tfm > 1``, ``imgs=data['img']`` returns a list with ``k_tfm`` image tensors. +- [May 2020] Added the person attribute recognition code used in `Omni-Scale Feature Learning for Person Re-Identification (ICCV'19) `_. See ``projects/attribute_recognition/``. +- [May 2020] ``v1.2.1``: Added a simple API for feature extraction (``torchreid/utils/feature_extractor.py``). See the `documentation `_ for the instruction. +- [Apr 2020] Code for reproducing the experiments of `deep mutual learning `_ in the `OSNet paper `__ (Supp. B) has been released at ``projects/DML``. +- [Apr 2020] Upgraded to ``v1.2.0``. The engine class has been made more model-agnostic to improve extensibility. See `Engine `_ and `ImageSoftmaxEngine `_ for more details. Credit to `Dassl.pytorch `_. +- [Dec 2019] Our `OSNet paper `_ has been updated, with additional experiments (in section B of the supplementary) showing some useful techniques for improving OSNet's performance in practice. +- [Nov 2019] ``ImageDataManager`` can load training data from target datasets by setting ``load_train_targets=True``, and the train-loader can be accessed with ``train_loader_t = datamanager.train_loader_t``. This feature is useful for domain adaptation research. + + +Installation +--------------- + +Make sure `conda `_ is installed. + + +.. code-block:: bash + + # cd to your preferred directory and clone this repo + git clone https://github.com/KaiyangZhou/deep-person-reid.git + + # create environment + cd deep-person-reid/ + conda create --name torchreid python=3.7 + conda activate torchreid + + # install dependencies + # make sure `which python` and `which pip` point to the correct path + pip install -r requirements.txt + + # install torch and torchvision (select the proper cuda version to suit your machine) + conda install pytorch torchvision cudatoolkit=9.0 -c pytorch + + # install torchreid (don't need to re-build it if you modify the source code) + python setup.py develop + + +Get started: 30 seconds to Torchreid +------------------------------------- +1. Import ``torchreid`` + +.. code-block:: python + + import torchreid + +2. Load data manager + +.. code-block:: python + + datamanager = torchreid.data.ImageDataManager( + root="reid-data", + sources="market1501", + targets="market1501", + height=256, + width=128, + batch_size_train=32, + batch_size_test=100, + transforms=["random_flip", "random_crop"] + ) + +3 Build model, optimizer and lr_scheduler + +.. code-block:: python + + model = torchreid.models.build_model( + name="resnet50", + num_classes=datamanager.num_train_pids, + loss="softmax", + pretrained=True + ) + + model = model.cuda() + + optimizer = torchreid.optim.build_optimizer( + model, + optim="adam", + lr=0.0003 + ) + + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler="single_step", + stepsize=20 + ) + +4. Build engine + +.. code-block:: python + + engine = torchreid.engine.ImageSoftmaxEngine( + datamanager, + model, + optimizer=optimizer, + scheduler=scheduler, + label_smooth=True + ) + +5. Run training and test + +.. code-block:: python + + engine.run( + save_dir="log/resnet50", + max_epoch=60, + eval_freq=10, + print_freq=10, + test_only=False + ) + + +A unified interface +----------------------- +In "deep-person-reid/scripts/", we provide a unified interface to train and test a model. See "scripts/main.py" and "scripts/default_config.py" for more details. The folder "configs/" contains some predefined configs which you can use as a starting point. + +Below we provide an example to train and test `OSNet (Zhou et al. ICCV'19) `_. Assume :code:`PATH_TO_DATA` is the directory containing reid datasets. The environmental variable :code:`CUDA_VISIBLE_DEVICES` is omitted, which you need to specify if you have a pool of gpus and want to use a specific set of them. + +Conventional setting +^^^^^^^^^^^^^^^^^^^^^ + +To train OSNet on Market1501, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA + + +The config file sets Market1501 as the default dataset. If you wanna use DukeMTMC-reID, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + -s dukemtmcreid \ + -t dukemtmcreid \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA \ + data.save_dir log/osnet_x1_0_dukemtmcreid_softmax_cosinelr + +The code will automatically (download and) load the ImageNet pretrained weights. After the training is done, the model will be saved as "log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250". Under the same folder, you can find the `tensorboard `_ file. To visualize the learning curves using tensorboard, you can run :code:`tensorboard --logdir=log/osnet_x1_0_market1501_softmax_cosinelr` in the terminal and visit :code:`http://localhost:6006/` in your web browser. + +Evaluation is automatically performed at the end of training. To run the test again using the trained model, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --root $PATH_TO_DATA \ + model.load_weights log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250 \ + test.evaluate True + + +Cross-domain setting +^^^^^^^^^^^^^^^^^^^^^ + +Suppose you wanna train OSNet on DukeMTMC-reID and test its performance on Market1501, you can do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad.yaml \ + -s dukemtmcreid \ + -t market1501 \ + --transforms random_flip color_jitter \ + --root $PATH_TO_DATA + +Here we only test the cross-domain performance. However, if you also want to test the performance on the source dataset, i.e. DukeMTMC-reID, you can set :code:`-t dukemtmcreid market1501`, which will evaluate the model on the two datasets separately. + +Different from the same-domain setting, here we replace :code:`random_erase` with :code:`color_jitter`. This can improve the generalization performance on the unseen target dataset. + +Pretrained models are available in the `Model Zoo `_. + + +Datasets +-------- + +Image-reid datasets +^^^^^^^^^^^^^^^^^^^^^ +- `Market1501 `_ +- `CUHK03 `_ +- `DukeMTMC-reID `_ +- `MSMT17 `_ +- `VIPeR `_ +- `GRID `_ +- `CUHK01 `_ +- `SenseReID `_ +- `QMUL-iLIDS `_ +- `PRID `_ + +Geo-localization datasets +^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `University-1652 `_ + +Video-reid datasets +^^^^^^^^^^^^^^^^^^^^^^^ +- `MARS `_ +- `iLIDS-VID `_ +- `PRID2011 `_ +- `DukeMTMC-VideoReID `_ + + +Models +------- + +ImageNet classification models +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `ResNet `_ +- `ResNeXt `_ +- `SENet `_ +- `DenseNet `_ +- `Inception-ResNet-V2 `_ +- `Inception-V4 `_ +- `Xception `_ +- `IBN-Net `_ + +Lightweight models +^^^^^^^^^^^^^^^^^^^ +- `NASNet `_ +- `MobileNetV2 `_ +- `ShuffleNet `_ +- `ShuffleNetV2 `_ +- `SqueezeNet `_ + +ReID-specific models +^^^^^^^^^^^^^^^^^^^^^^ +- `MuDeep `_ +- `ResNet-mid `_ +- `HACNN `_ +- `PCB `_ +- `MLFN `_ +- `OSNet `_ +- `OSNet-AIN `_ + + +Useful links +------------- +- `OSNet-IBN1-Lite (test-only code with lite docker container) `_ +- `Deep Learning for Person Re-identification: A Survey and Outlook `_ + + +Citation +--------- +If you use this code or the models in your research, please give credit to the following papers: + +.. code-block:: bash + + @article{torchreid, + title={Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch}, + author={Zhou, Kaiyang and Xiang, Tao}, + journal={arXiv preprint arXiv:1910.10093}, + year={2019} + } + + @inproceedings{zhou2019osnet, + title={Omni-Scale Feature Learning for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + booktitle={ICCV}, + year={2019} + } + + @article{zhou2021osnet, + title={Learning Generalisable Omni-Scale Representations for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + journal={TPAMI}, + year={2021} + } + + diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt new file mode 100644 index 0000000000..9adb71925d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt @@ -0,0 +1,87 @@ +LICENSE +README.rst +setup.py +torchreid/__init__.py +torchreid.egg-info/PKG-INFO +torchreid.egg-info/SOURCES.txt +torchreid.egg-info/dependency_links.txt +torchreid.egg-info/requires.txt +torchreid.egg-info/top_level.txt +torchreid/data/__init__.py +torchreid/data/datamanager.py +torchreid/data/sampler.py +torchreid/data/transforms.py +torchreid/data/datasets/__init__.py +torchreid/data/datasets/dataset.py +torchreid/data/datasets/image/__init__.py +torchreid/data/datasets/image/cuhk01.py +torchreid/data/datasets/image/cuhk02.py +torchreid/data/datasets/image/cuhk03.py +torchreid/data/datasets/image/cuhksysu.py +torchreid/data/datasets/image/dukemtmcreid.py +torchreid/data/datasets/image/grid.py +torchreid/data/datasets/image/ilids.py +torchreid/data/datasets/image/market1501.py +torchreid/data/datasets/image/msmt17.py +torchreid/data/datasets/image/prid.py +torchreid/data/datasets/image/sensereid.py +torchreid/data/datasets/image/university1652.py +torchreid/data/datasets/image/viper.py +torchreid/data/datasets/video/__init__.py +torchreid/data/datasets/video/dukemtmcvidreid.py +torchreid/data/datasets/video/ilidsvid.py +torchreid/data/datasets/video/mars.py +torchreid/data/datasets/video/prid2011.py +torchreid/engine/__init__.py +torchreid/engine/engine.py +torchreid/engine/image/__init__.py +torchreid/engine/image/softmax.py +torchreid/engine/image/triplet.py +torchreid/engine/video/__init__.py +torchreid/engine/video/softmax.py +torchreid/engine/video/triplet.py +torchreid/losses/__init__.py +torchreid/losses/cross_entropy_loss.py +torchreid/losses/hard_mine_triplet_loss.py +torchreid/metrics/__init__.py +torchreid/metrics/accuracy.py +torchreid/metrics/distance.py +torchreid/metrics/rank.py +torchreid/metrics/rank_cylib/__init__.py +torchreid/metrics/rank_cylib/rank_cy.c +torchreid/metrics/rank_cylib/setup.py +torchreid/metrics/rank_cylib/test_cython.py +torchreid/models/__init__.py +torchreid/models/densenet.py +torchreid/models/hacnn.py +torchreid/models/inceptionresnetv2.py +torchreid/models/inceptionv4.py +torchreid/models/mlfn.py +torchreid/models/mobilenetv2.py +torchreid/models/mudeep.py +torchreid/models/nasnet.py +torchreid/models/osnet.py +torchreid/models/osnet_ain.py +torchreid/models/pcb.py +torchreid/models/resnet.py +torchreid/models/resnet_ibn_a.py +torchreid/models/resnet_ibn_b.py +torchreid/models/resnetmid.py +torchreid/models/senet.py +torchreid/models/shufflenet.py +torchreid/models/shufflenetv2.py +torchreid/models/squeezenet.py +torchreid/models/xception.py +torchreid/optim/__init__.py +torchreid/optim/lr_scheduler.py +torchreid/optim/optimizer.py +torchreid/optim/radam.py +torchreid/utils/__init__.py +torchreid/utils/avgmeter.py +torchreid/utils/feature_extractor.py +torchreid/utils/loggers.py +torchreid/utils/model_complexity.py +torchreid/utils/reidtools.py +torchreid/utils/rerank.py +torchreid/utils/tools.py +torchreid/utils/torchtools.py \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt new file mode 100644 index 0000000000..8b13789179 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt @@ -0,0 +1 @@ + diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt new file mode 100644 index 0000000000..bb0003a987 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt @@ -0,0 +1,16 @@ +numpy +Cython +h5py +Pillow +six +scipy +opencv-python +matplotlib +tb-nightly +future +yacs +gdown +flake8 +yapf +isort +imageio diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt new file mode 100644 index 0000000000..a3cd384f80 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt @@ -0,0 +1 @@ +torchreid -- Gitee From b537c2c96bd51690a4a1480a3a636182659da9d6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:49:16 +0000 Subject: [PATCH 03/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/torchreid.egg-info?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../OSNet/torchreid.egg-info/.keep | 0 .../torchreid.egg-info/PKG-INFO | 330 ------------------ .../torchreid.egg-info/SOURCES.txt | 87 ----- .../torchreid.egg-info/dependency_links.txt | 1 - .../torchreid.egg-info/requires.txt | 16 - .../torchreid.egg-info/top_level.txt | 1 - 6 files changed, 435 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/.keep deleted file mode 100644 index e69de29bb2..0000000000 diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO deleted file mode 100644 index ea44b1cca6..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/PKG-INFO +++ /dev/null @@ -1,330 +0,0 @@ -Metadata-Version: 2.1 -Name: torchreid -Version: 1.4.0 -Summary: A library for deep learning person re-ID in PyTorch -Home-page: https://github.com/KaiyangZhou/deep-person-reid -Author: Kaiyang Zhou -License: MIT -Keywords: Person Re-Identification,Deep Learning,Computer Vision -Platform: UNKNOWN -License-File: LICENSE - -Torchreid -=========== -Torchreid is a library for deep-learning person re-identification, written in `PyTorch `_ and developed for our ICCV'19 project, `Omni-Scale Feature Learning for Person Re-Identification `_. - -It features: - -- multi-GPU training -- support both image- and video-reid -- end-to-end training and evaluation -- incredibly easy preparation of reid datasets -- multi-dataset training -- cross-dataset evaluation -- standard protocol used by most research papers -- highly extensible (easy to add models, datasets, training methods, etc.) -- implementations of state-of-the-art deep reid models -- access to pretrained reid models -- advanced training techniques -- visualization tools (tensorboard, ranks, etc.) - - -Code: https://github.com/KaiyangZhou/deep-person-reid. - -Documentation: https://kaiyangzhou.github.io/deep-person-reid/. - -How-to instructions: https://kaiyangzhou.github.io/deep-person-reid/user_guide. - -Model zoo: https://kaiyangzhou.github.io/deep-person-reid/MODEL_ZOO. - -Tech report: https://arxiv.org/abs/1910.10093. - -You can find some research projects that are built on top of Torchreid `here `_. - - -What's new ------------- -- [Aug 2021] We have released the ImageNet-pretrained models of ``osnet_ain_x0_75``, ``osnet_ain_x0_5`` and ``osnet_ain_x0_25``. The pretraining setup follows `pycls `_. -- [Apr 2021] We have updated the appendix in the `TPAMI version of OSNet `_ to include results in the multi-source domain generalization setting. The trained models can be found in the `Model Zoo `_. -- [Apr 2021] We have added a script to automate the process of calculating average results over multiple splits. For more details please see ``tools/parse_test_res.py``. -- [Apr 2021] ``v1.4.0``: We added the person search dataset, `CUHK-SYSU `_. Please see the `documentation `_ regarding how to download the dataset (it contains cropped person images). -- [Apr 2021] All models in the model zoo have been moved to google drive. Please raise an issue if any model's performance is inconsistent with the numbers shown in the model zoo page (could be caused by wrong links). -- [Mar 2021] `OSNet `_ will appear in the TPAMI journal! Compared with the conference version, which focuses on discriminative feature learning using the omni-scale building block, this journal extension further considers generalizable feature learning by integrating `instance normalization layers `_ with the OSNet architecture. We hope this journal paper can motivate more future work to taclke the generalization issue in cross-dataset re-ID. -- [Mar 2021] Generalization across domains (datasets) in person re-ID is crucial in real-world applications, which is closely related to the topic of *domain generalization*. Interested in learning how the field of domain generalization has developed over the last decade? Check our recent survey in this topic at https://arxiv.org/abs/2103.02503, with coverage on the history, datasets, related problems, methodologies, potential directions, and so on (*methods designed for generalizable re-ID are also covered*!). -- [Feb 2021] ``v1.3.6`` Added `University-1652 `_, a new dataset for multi-view multi-source geo-localization (credit to `Zhedong Zheng `_). -- [Feb 2021] ``v1.3.5``: Now the `cython code `_ works on Windows (credit to `lablabla `_). -- [Jan 2021] Our recent work, `MixStyle `_ (mixing instance-level feature statistics of samples of different domains for improving domain generalization), has been accepted to ICLR'21. The code has been released at https://github.com/KaiyangZhou/mixstyle-release where the person re-ID part is based on Torchreid. -- [Jan 2021] A new evaluation metric called `mean Inverse Negative Penalty (mINP)` for person re-ID has been introduced in `Deep Learning for Person Re-identification: A Survey and Outlook (TPAMI 2021) `_. Their code can be accessed at ``_. -- [Aug 2020] ``v1.3.3``: Fixed bug in ``visrank`` (caused by not unpacking ``dsetid``). -- [Aug 2020] ``v1.3.2``: Added ``_junk_pids`` to ``grid`` and ``prid``. This avoids using mislabeled gallery images for training when setting ``combineall=True``. -- [Aug 2020] ``v1.3.0``: (1) Added ``dsetid`` to the existing 3-tuple data source, resulting in ``(impath, pid, camid, dsetid)``. This variable denotes the dataset ID and is useful when combining multiple datasets for training (as a dataset indicator). E.g., when combining ``market1501`` and ``cuhk03``, the former will be assigned ``dsetid=0`` while the latter will be assigned ``dsetid=1``. (2) Added ``RandomDatasetSampler``. Analogous to ``RandomDomainSampler``, ``RandomDatasetSampler`` samples a certain number of images (``batch_size // num_datasets``) from each of specified datasets (the amount is determined by ``num_datasets``). -- [Aug 2020] ``v1.2.6``: Added ``RandomDomainSampler`` (it samples ``num_cams`` cameras each with ``batch_size // num_cams`` images to form a mini-batch). -- [Jun 2020] ``v1.2.5``: (1) Dataloader's output from ``__getitem__`` has been changed from ``list`` to ``dict``. Previously, an element, e.g. image tensor, was fetched with ``imgs=data[0]``. Now it should be obtained by ``imgs=data['img']``. See this `commit `_ for detailed changes. (2) Added ``k_tfm`` as an option to image data loader, which allows data augmentation to be applied ``k_tfm`` times *independently* to an image. If ``k_tfm > 1``, ``imgs=data['img']`` returns a list with ``k_tfm`` image tensors. -- [May 2020] Added the person attribute recognition code used in `Omni-Scale Feature Learning for Person Re-Identification (ICCV'19) `_. See ``projects/attribute_recognition/``. -- [May 2020] ``v1.2.1``: Added a simple API for feature extraction (``torchreid/utils/feature_extractor.py``). See the `documentation `_ for the instruction. -- [Apr 2020] Code for reproducing the experiments of `deep mutual learning `_ in the `OSNet paper `__ (Supp. B) has been released at ``projects/DML``. -- [Apr 2020] Upgraded to ``v1.2.0``. The engine class has been made more model-agnostic to improve extensibility. See `Engine `_ and `ImageSoftmaxEngine `_ for more details. Credit to `Dassl.pytorch `_. -- [Dec 2019] Our `OSNet paper `_ has been updated, with additional experiments (in section B of the supplementary) showing some useful techniques for improving OSNet's performance in practice. -- [Nov 2019] ``ImageDataManager`` can load training data from target datasets by setting ``load_train_targets=True``, and the train-loader can be accessed with ``train_loader_t = datamanager.train_loader_t``. This feature is useful for domain adaptation research. - - -Installation ---------------- - -Make sure `conda `_ is installed. - - -.. code-block:: bash - - # cd to your preferred directory and clone this repo - git clone https://github.com/KaiyangZhou/deep-person-reid.git - - # create environment - cd deep-person-reid/ - conda create --name torchreid python=3.7 - conda activate torchreid - - # install dependencies - # make sure `which python` and `which pip` point to the correct path - pip install -r requirements.txt - - # install torch and torchvision (select the proper cuda version to suit your machine) - conda install pytorch torchvision cudatoolkit=9.0 -c pytorch - - # install torchreid (don't need to re-build it if you modify the source code) - python setup.py develop - - -Get started: 30 seconds to Torchreid -------------------------------------- -1. Import ``torchreid`` - -.. code-block:: python - - import torchreid - -2. Load data manager - -.. code-block:: python - - datamanager = torchreid.data.ImageDataManager( - root="reid-data", - sources="market1501", - targets="market1501", - height=256, - width=128, - batch_size_train=32, - batch_size_test=100, - transforms=["random_flip", "random_crop"] - ) - -3 Build model, optimizer and lr_scheduler - -.. code-block:: python - - model = torchreid.models.build_model( - name="resnet50", - num_classes=datamanager.num_train_pids, - loss="softmax", - pretrained=True - ) - - model = model.cuda() - - optimizer = torchreid.optim.build_optimizer( - model, - optim="adam", - lr=0.0003 - ) - - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler="single_step", - stepsize=20 - ) - -4. Build engine - -.. code-block:: python - - engine = torchreid.engine.ImageSoftmaxEngine( - datamanager, - model, - optimizer=optimizer, - scheduler=scheduler, - label_smooth=True - ) - -5. Run training and test - -.. code-block:: python - - engine.run( - save_dir="log/resnet50", - max_epoch=60, - eval_freq=10, - print_freq=10, - test_only=False - ) - - -A unified interface ------------------------ -In "deep-person-reid/scripts/", we provide a unified interface to train and test a model. See "scripts/main.py" and "scripts/default_config.py" for more details. The folder "configs/" contains some predefined configs which you can use as a starting point. - -Below we provide an example to train and test `OSNet (Zhou et al. ICCV'19) `_. Assume :code:`PATH_TO_DATA` is the directory containing reid datasets. The environmental variable :code:`CUDA_VISIBLE_DEVICES` is omitted, which you need to specify if you have a pool of gpus and want to use a specific set of them. - -Conventional setting -^^^^^^^^^^^^^^^^^^^^^ - -To train OSNet on Market1501, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - --transforms random_flip random_erase \ - --root $PATH_TO_DATA - - -The config file sets Market1501 as the default dataset. If you wanna use DukeMTMC-reID, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - -s dukemtmcreid \ - -t dukemtmcreid \ - --transforms random_flip random_erase \ - --root $PATH_TO_DATA \ - data.save_dir log/osnet_x1_0_dukemtmcreid_softmax_cosinelr - -The code will automatically (download and) load the ImageNet pretrained weights. After the training is done, the model will be saved as "log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250". Under the same folder, you can find the `tensorboard `_ file. To visualize the learning curves using tensorboard, you can run :code:`tensorboard --logdir=log/osnet_x1_0_market1501_softmax_cosinelr` in the terminal and visit :code:`http://localhost:6006/` in your web browser. - -Evaluation is automatically performed at the end of training. To run the test again using the trained model, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - --root $PATH_TO_DATA \ - model.load_weights log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250 \ - test.evaluate True - - -Cross-domain setting -^^^^^^^^^^^^^^^^^^^^^ - -Suppose you wanna train OSNet on DukeMTMC-reID and test its performance on Market1501, you can do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad.yaml \ - -s dukemtmcreid \ - -t market1501 \ - --transforms random_flip color_jitter \ - --root $PATH_TO_DATA - -Here we only test the cross-domain performance. However, if you also want to test the performance on the source dataset, i.e. DukeMTMC-reID, you can set :code:`-t dukemtmcreid market1501`, which will evaluate the model on the two datasets separately. - -Different from the same-domain setting, here we replace :code:`random_erase` with :code:`color_jitter`. This can improve the generalization performance on the unseen target dataset. - -Pretrained models are available in the `Model Zoo `_. - - -Datasets --------- - -Image-reid datasets -^^^^^^^^^^^^^^^^^^^^^ -- `Market1501 `_ -- `CUHK03 `_ -- `DukeMTMC-reID `_ -- `MSMT17 `_ -- `VIPeR `_ -- `GRID `_ -- `CUHK01 `_ -- `SenseReID `_ -- `QMUL-iLIDS `_ -- `PRID `_ - -Geo-localization datasets -^^^^^^^^^^^^^^^^^^^^^^^^^^^ -- `University-1652 `_ - -Video-reid datasets -^^^^^^^^^^^^^^^^^^^^^^^ -- `MARS `_ -- `iLIDS-VID `_ -- `PRID2011 `_ -- `DukeMTMC-VideoReID `_ - - -Models -------- - -ImageNet classification models -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -- `ResNet `_ -- `ResNeXt `_ -- `SENet `_ -- `DenseNet `_ -- `Inception-ResNet-V2 `_ -- `Inception-V4 `_ -- `Xception `_ -- `IBN-Net `_ - -Lightweight models -^^^^^^^^^^^^^^^^^^^ -- `NASNet `_ -- `MobileNetV2 `_ -- `ShuffleNet `_ -- `ShuffleNetV2 `_ -- `SqueezeNet `_ - -ReID-specific models -^^^^^^^^^^^^^^^^^^^^^^ -- `MuDeep `_ -- `ResNet-mid `_ -- `HACNN `_ -- `PCB `_ -- `MLFN `_ -- `OSNet `_ -- `OSNet-AIN `_ - - -Useful links -------------- -- `OSNet-IBN1-Lite (test-only code with lite docker container) `_ -- `Deep Learning for Person Re-identification: A Survey and Outlook `_ - - -Citation ---------- -If you use this code or the models in your research, please give credit to the following papers: - -.. code-block:: bash - - @article{torchreid, - title={Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch}, - author={Zhou, Kaiyang and Xiang, Tao}, - journal={arXiv preprint arXiv:1910.10093}, - year={2019} - } - - @inproceedings{zhou2019osnet, - title={Omni-Scale Feature Learning for Person Re-Identification}, - author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, - booktitle={ICCV}, - year={2019} - } - - @article{zhou2021osnet, - title={Learning Generalisable Omni-Scale Representations for Person Re-Identification}, - author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, - journal={TPAMI}, - year={2021} - } - - diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt deleted file mode 100644 index 9adb71925d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/SOURCES.txt +++ /dev/null @@ -1,87 +0,0 @@ -LICENSE -README.rst -setup.py -torchreid/__init__.py -torchreid.egg-info/PKG-INFO -torchreid.egg-info/SOURCES.txt -torchreid.egg-info/dependency_links.txt -torchreid.egg-info/requires.txt -torchreid.egg-info/top_level.txt -torchreid/data/__init__.py -torchreid/data/datamanager.py -torchreid/data/sampler.py -torchreid/data/transforms.py -torchreid/data/datasets/__init__.py -torchreid/data/datasets/dataset.py -torchreid/data/datasets/image/__init__.py -torchreid/data/datasets/image/cuhk01.py -torchreid/data/datasets/image/cuhk02.py -torchreid/data/datasets/image/cuhk03.py -torchreid/data/datasets/image/cuhksysu.py -torchreid/data/datasets/image/dukemtmcreid.py -torchreid/data/datasets/image/grid.py -torchreid/data/datasets/image/ilids.py -torchreid/data/datasets/image/market1501.py -torchreid/data/datasets/image/msmt17.py -torchreid/data/datasets/image/prid.py -torchreid/data/datasets/image/sensereid.py -torchreid/data/datasets/image/university1652.py -torchreid/data/datasets/image/viper.py -torchreid/data/datasets/video/__init__.py -torchreid/data/datasets/video/dukemtmcvidreid.py -torchreid/data/datasets/video/ilidsvid.py -torchreid/data/datasets/video/mars.py -torchreid/data/datasets/video/prid2011.py -torchreid/engine/__init__.py -torchreid/engine/engine.py -torchreid/engine/image/__init__.py -torchreid/engine/image/softmax.py -torchreid/engine/image/triplet.py -torchreid/engine/video/__init__.py -torchreid/engine/video/softmax.py -torchreid/engine/video/triplet.py -torchreid/losses/__init__.py -torchreid/losses/cross_entropy_loss.py -torchreid/losses/hard_mine_triplet_loss.py -torchreid/metrics/__init__.py -torchreid/metrics/accuracy.py -torchreid/metrics/distance.py -torchreid/metrics/rank.py -torchreid/metrics/rank_cylib/__init__.py -torchreid/metrics/rank_cylib/rank_cy.c -torchreid/metrics/rank_cylib/setup.py -torchreid/metrics/rank_cylib/test_cython.py -torchreid/models/__init__.py -torchreid/models/densenet.py -torchreid/models/hacnn.py -torchreid/models/inceptionresnetv2.py -torchreid/models/inceptionv4.py -torchreid/models/mlfn.py -torchreid/models/mobilenetv2.py -torchreid/models/mudeep.py -torchreid/models/nasnet.py -torchreid/models/osnet.py -torchreid/models/osnet_ain.py -torchreid/models/pcb.py -torchreid/models/resnet.py -torchreid/models/resnet_ibn_a.py -torchreid/models/resnet_ibn_b.py -torchreid/models/resnetmid.py -torchreid/models/senet.py -torchreid/models/shufflenet.py -torchreid/models/shufflenetv2.py -torchreid/models/squeezenet.py -torchreid/models/xception.py -torchreid/optim/__init__.py -torchreid/optim/lr_scheduler.py -torchreid/optim/optimizer.py -torchreid/optim/radam.py -torchreid/utils/__init__.py -torchreid/utils/avgmeter.py -torchreid/utils/feature_extractor.py -torchreid/utils/loggers.py -torchreid/utils/model_complexity.py -torchreid/utils/reidtools.py -torchreid/utils/rerank.py -torchreid/utils/tools.py -torchreid/utils/torchtools.py \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt deleted file mode 100644 index 8b13789179..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/dependency_links.txt +++ /dev/null @@ -1 +0,0 @@ - diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt deleted file mode 100644 index bb0003a987..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/requires.txt +++ /dev/null @@ -1,16 +0,0 @@ -numpy -Cython -h5py -Pillow -six -scipy -opencv-python -matplotlib -tb-nightly -future -yacs -gdown -flake8 -yapf -isort -imageio diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt deleted file mode 100644 index a3cd384f80..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/torchreid.egg-info/top_level.txt +++ /dev/null @@ -1 +0,0 @@ -torchreid -- Gitee From 85359d8f54aec8a1322238489130f23ca8bcba01 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:49:33 +0000 Subject: [PATCH 04/31] init --- .../OSNet/torchreid.egg-info/PKG-INFO | 330 ++++++++++++++++++ .../OSNet/torchreid.egg-info/SOURCES.txt | 87 +++++ .../torchreid.egg-info/dependency_links.txt | 1 + .../OSNet/torchreid.egg-info/requires.txt | 16 + .../OSNet/torchreid.egg-info/top_level.txt | 1 + 5 files changed, 435 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/PKG-INFO create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/SOURCES.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/dependency_links.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/requires.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/top_level.txt diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/PKG-INFO b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/PKG-INFO new file mode 100644 index 0000000000..ea44b1cca6 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/PKG-INFO @@ -0,0 +1,330 @@ +Metadata-Version: 2.1 +Name: torchreid +Version: 1.4.0 +Summary: A library for deep learning person re-ID in PyTorch +Home-page: https://github.com/KaiyangZhou/deep-person-reid +Author: Kaiyang Zhou +License: MIT +Keywords: Person Re-Identification,Deep Learning,Computer Vision +Platform: UNKNOWN +License-File: LICENSE + +Torchreid +=========== +Torchreid is a library for deep-learning person re-identification, written in `PyTorch `_ and developed for our ICCV'19 project, `Omni-Scale Feature Learning for Person Re-Identification `_. + +It features: + +- multi-GPU training +- support both image- and video-reid +- end-to-end training and evaluation +- incredibly easy preparation of reid datasets +- multi-dataset training +- cross-dataset evaluation +- standard protocol used by most research papers +- highly extensible (easy to add models, datasets, training methods, etc.) +- implementations of state-of-the-art deep reid models +- access to pretrained reid models +- advanced training techniques +- visualization tools (tensorboard, ranks, etc.) + + +Code: https://github.com/KaiyangZhou/deep-person-reid. + +Documentation: https://kaiyangzhou.github.io/deep-person-reid/. + +How-to instructions: https://kaiyangzhou.github.io/deep-person-reid/user_guide. + +Model zoo: https://kaiyangzhou.github.io/deep-person-reid/MODEL_ZOO. + +Tech report: https://arxiv.org/abs/1910.10093. + +You can find some research projects that are built on top of Torchreid `here `_. + + +What's new +------------ +- [Aug 2021] We have released the ImageNet-pretrained models of ``osnet_ain_x0_75``, ``osnet_ain_x0_5`` and ``osnet_ain_x0_25``. The pretraining setup follows `pycls `_. +- [Apr 2021] We have updated the appendix in the `TPAMI version of OSNet `_ to include results in the multi-source domain generalization setting. The trained models can be found in the `Model Zoo `_. +- [Apr 2021] We have added a script to automate the process of calculating average results over multiple splits. For more details please see ``tools/parse_test_res.py``. +- [Apr 2021] ``v1.4.0``: We added the person search dataset, `CUHK-SYSU `_. Please see the `documentation `_ regarding how to download the dataset (it contains cropped person images). +- [Apr 2021] All models in the model zoo have been moved to google drive. Please raise an issue if any model's performance is inconsistent with the numbers shown in the model zoo page (could be caused by wrong links). +- [Mar 2021] `OSNet `_ will appear in the TPAMI journal! Compared with the conference version, which focuses on discriminative feature learning using the omni-scale building block, this journal extension further considers generalizable feature learning by integrating `instance normalization layers `_ with the OSNet architecture. We hope this journal paper can motivate more future work to taclke the generalization issue in cross-dataset re-ID. +- [Mar 2021] Generalization across domains (datasets) in person re-ID is crucial in real-world applications, which is closely related to the topic of *domain generalization*. Interested in learning how the field of domain generalization has developed over the last decade? Check our recent survey in this topic at https://arxiv.org/abs/2103.02503, with coverage on the history, datasets, related problems, methodologies, potential directions, and so on (*methods designed for generalizable re-ID are also covered*!). +- [Feb 2021] ``v1.3.6`` Added `University-1652 `_, a new dataset for multi-view multi-source geo-localization (credit to `Zhedong Zheng `_). +- [Feb 2021] ``v1.3.5``: Now the `cython code `_ works on Windows (credit to `lablabla `_). +- [Jan 2021] Our recent work, `MixStyle `_ (mixing instance-level feature statistics of samples of different domains for improving domain generalization), has been accepted to ICLR'21. The code has been released at https://github.com/KaiyangZhou/mixstyle-release where the person re-ID part is based on Torchreid. +- [Jan 2021] A new evaluation metric called `mean Inverse Negative Penalty (mINP)` for person re-ID has been introduced in `Deep Learning for Person Re-identification: A Survey and Outlook (TPAMI 2021) `_. Their code can be accessed at ``_. +- [Aug 2020] ``v1.3.3``: Fixed bug in ``visrank`` (caused by not unpacking ``dsetid``). +- [Aug 2020] ``v1.3.2``: Added ``_junk_pids`` to ``grid`` and ``prid``. This avoids using mislabeled gallery images for training when setting ``combineall=True``. +- [Aug 2020] ``v1.3.0``: (1) Added ``dsetid`` to the existing 3-tuple data source, resulting in ``(impath, pid, camid, dsetid)``. This variable denotes the dataset ID and is useful when combining multiple datasets for training (as a dataset indicator). E.g., when combining ``market1501`` and ``cuhk03``, the former will be assigned ``dsetid=0`` while the latter will be assigned ``dsetid=1``. (2) Added ``RandomDatasetSampler``. Analogous to ``RandomDomainSampler``, ``RandomDatasetSampler`` samples a certain number of images (``batch_size // num_datasets``) from each of specified datasets (the amount is determined by ``num_datasets``). +- [Aug 2020] ``v1.2.6``: Added ``RandomDomainSampler`` (it samples ``num_cams`` cameras each with ``batch_size // num_cams`` images to form a mini-batch). +- [Jun 2020] ``v1.2.5``: (1) Dataloader's output from ``__getitem__`` has been changed from ``list`` to ``dict``. Previously, an element, e.g. image tensor, was fetched with ``imgs=data[0]``. Now it should be obtained by ``imgs=data['img']``. See this `commit `_ for detailed changes. (2) Added ``k_tfm`` as an option to image data loader, which allows data augmentation to be applied ``k_tfm`` times *independently* to an image. If ``k_tfm > 1``, ``imgs=data['img']`` returns a list with ``k_tfm`` image tensors. +- [May 2020] Added the person attribute recognition code used in `Omni-Scale Feature Learning for Person Re-Identification (ICCV'19) `_. See ``projects/attribute_recognition/``. +- [May 2020] ``v1.2.1``: Added a simple API for feature extraction (``torchreid/utils/feature_extractor.py``). See the `documentation `_ for the instruction. +- [Apr 2020] Code for reproducing the experiments of `deep mutual learning `_ in the `OSNet paper `__ (Supp. B) has been released at ``projects/DML``. +- [Apr 2020] Upgraded to ``v1.2.0``. The engine class has been made more model-agnostic to improve extensibility. See `Engine `_ and `ImageSoftmaxEngine `_ for more details. Credit to `Dassl.pytorch `_. +- [Dec 2019] Our `OSNet paper `_ has been updated, with additional experiments (in section B of the supplementary) showing some useful techniques for improving OSNet's performance in practice. +- [Nov 2019] ``ImageDataManager`` can load training data from target datasets by setting ``load_train_targets=True``, and the train-loader can be accessed with ``train_loader_t = datamanager.train_loader_t``. This feature is useful for domain adaptation research. + + +Installation +--------------- + +Make sure `conda `_ is installed. + + +.. code-block:: bash + + # cd to your preferred directory and clone this repo + git clone https://github.com/KaiyangZhou/deep-person-reid.git + + # create environment + cd deep-person-reid/ + conda create --name torchreid python=3.7 + conda activate torchreid + + # install dependencies + # make sure `which python` and `which pip` point to the correct path + pip install -r requirements.txt + + # install torch and torchvision (select the proper cuda version to suit your machine) + conda install pytorch torchvision cudatoolkit=9.0 -c pytorch + + # install torchreid (don't need to re-build it if you modify the source code) + python setup.py develop + + +Get started: 30 seconds to Torchreid +------------------------------------- +1. Import ``torchreid`` + +.. code-block:: python + + import torchreid + +2. Load data manager + +.. code-block:: python + + datamanager = torchreid.data.ImageDataManager( + root="reid-data", + sources="market1501", + targets="market1501", + height=256, + width=128, + batch_size_train=32, + batch_size_test=100, + transforms=["random_flip", "random_crop"] + ) + +3 Build model, optimizer and lr_scheduler + +.. code-block:: python + + model = torchreid.models.build_model( + name="resnet50", + num_classes=datamanager.num_train_pids, + loss="softmax", + pretrained=True + ) + + model = model.cuda() + + optimizer = torchreid.optim.build_optimizer( + model, + optim="adam", + lr=0.0003 + ) + + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler="single_step", + stepsize=20 + ) + +4. Build engine + +.. code-block:: python + + engine = torchreid.engine.ImageSoftmaxEngine( + datamanager, + model, + optimizer=optimizer, + scheduler=scheduler, + label_smooth=True + ) + +5. Run training and test + +.. code-block:: python + + engine.run( + save_dir="log/resnet50", + max_epoch=60, + eval_freq=10, + print_freq=10, + test_only=False + ) + + +A unified interface +----------------------- +In "deep-person-reid/scripts/", we provide a unified interface to train and test a model. See "scripts/main.py" and "scripts/default_config.py" for more details. The folder "configs/" contains some predefined configs which you can use as a starting point. + +Below we provide an example to train and test `OSNet (Zhou et al. ICCV'19) `_. Assume :code:`PATH_TO_DATA` is the directory containing reid datasets. The environmental variable :code:`CUDA_VISIBLE_DEVICES` is omitted, which you need to specify if you have a pool of gpus and want to use a specific set of them. + +Conventional setting +^^^^^^^^^^^^^^^^^^^^^ + +To train OSNet on Market1501, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA + + +The config file sets Market1501 as the default dataset. If you wanna use DukeMTMC-reID, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + -s dukemtmcreid \ + -t dukemtmcreid \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA \ + data.save_dir log/osnet_x1_0_dukemtmcreid_softmax_cosinelr + +The code will automatically (download and) load the ImageNet pretrained weights. After the training is done, the model will be saved as "log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250". Under the same folder, you can find the `tensorboard `_ file. To visualize the learning curves using tensorboard, you can run :code:`tensorboard --logdir=log/osnet_x1_0_market1501_softmax_cosinelr` in the terminal and visit :code:`http://localhost:6006/` in your web browser. + +Evaluation is automatically performed at the end of training. To run the test again using the trained model, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --root $PATH_TO_DATA \ + model.load_weights log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250 \ + test.evaluate True + + +Cross-domain setting +^^^^^^^^^^^^^^^^^^^^^ + +Suppose you wanna train OSNet on DukeMTMC-reID and test its performance on Market1501, you can do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad.yaml \ + -s dukemtmcreid \ + -t market1501 \ + --transforms random_flip color_jitter \ + --root $PATH_TO_DATA + +Here we only test the cross-domain performance. However, if you also want to test the performance on the source dataset, i.e. DukeMTMC-reID, you can set :code:`-t dukemtmcreid market1501`, which will evaluate the model on the two datasets separately. + +Different from the same-domain setting, here we replace :code:`random_erase` with :code:`color_jitter`. This can improve the generalization performance on the unseen target dataset. + +Pretrained models are available in the `Model Zoo `_. + + +Datasets +-------- + +Image-reid datasets +^^^^^^^^^^^^^^^^^^^^^ +- `Market1501 `_ +- `CUHK03 `_ +- `DukeMTMC-reID `_ +- `MSMT17 `_ +- `VIPeR `_ +- `GRID `_ +- `CUHK01 `_ +- `SenseReID `_ +- `QMUL-iLIDS `_ +- `PRID `_ + +Geo-localization datasets +^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `University-1652 `_ + +Video-reid datasets +^^^^^^^^^^^^^^^^^^^^^^^ +- `MARS `_ +- `iLIDS-VID `_ +- `PRID2011 `_ +- `DukeMTMC-VideoReID `_ + + +Models +------- + +ImageNet classification models +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `ResNet `_ +- `ResNeXt `_ +- `SENet `_ +- `DenseNet `_ +- `Inception-ResNet-V2 `_ +- `Inception-V4 `_ +- `Xception `_ +- `IBN-Net `_ + +Lightweight models +^^^^^^^^^^^^^^^^^^^ +- `NASNet `_ +- `MobileNetV2 `_ +- `ShuffleNet `_ +- `ShuffleNetV2 `_ +- `SqueezeNet `_ + +ReID-specific models +^^^^^^^^^^^^^^^^^^^^^^ +- `MuDeep `_ +- `ResNet-mid `_ +- `HACNN `_ +- `PCB `_ +- `MLFN `_ +- `OSNet `_ +- `OSNet-AIN `_ + + +Useful links +------------- +- `OSNet-IBN1-Lite (test-only code with lite docker container) `_ +- `Deep Learning for Person Re-identification: A Survey and Outlook `_ + + +Citation +--------- +If you use this code or the models in your research, please give credit to the following papers: + +.. code-block:: bash + + @article{torchreid, + title={Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch}, + author={Zhou, Kaiyang and Xiang, Tao}, + journal={arXiv preprint arXiv:1910.10093}, + year={2019} + } + + @inproceedings{zhou2019osnet, + title={Omni-Scale Feature Learning for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + booktitle={ICCV}, + year={2019} + } + + @article{zhou2021osnet, + title={Learning Generalisable Omni-Scale Representations for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + journal={TPAMI}, + year={2021} + } + + diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/SOURCES.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/SOURCES.txt new file mode 100644 index 0000000000..9adb71925d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/SOURCES.txt @@ -0,0 +1,87 @@ +LICENSE +README.rst +setup.py +torchreid/__init__.py +torchreid.egg-info/PKG-INFO +torchreid.egg-info/SOURCES.txt +torchreid.egg-info/dependency_links.txt +torchreid.egg-info/requires.txt +torchreid.egg-info/top_level.txt +torchreid/data/__init__.py +torchreid/data/datamanager.py +torchreid/data/sampler.py +torchreid/data/transforms.py +torchreid/data/datasets/__init__.py +torchreid/data/datasets/dataset.py +torchreid/data/datasets/image/__init__.py +torchreid/data/datasets/image/cuhk01.py +torchreid/data/datasets/image/cuhk02.py +torchreid/data/datasets/image/cuhk03.py +torchreid/data/datasets/image/cuhksysu.py +torchreid/data/datasets/image/dukemtmcreid.py +torchreid/data/datasets/image/grid.py +torchreid/data/datasets/image/ilids.py +torchreid/data/datasets/image/market1501.py +torchreid/data/datasets/image/msmt17.py +torchreid/data/datasets/image/prid.py +torchreid/data/datasets/image/sensereid.py +torchreid/data/datasets/image/university1652.py +torchreid/data/datasets/image/viper.py +torchreid/data/datasets/video/__init__.py +torchreid/data/datasets/video/dukemtmcvidreid.py +torchreid/data/datasets/video/ilidsvid.py +torchreid/data/datasets/video/mars.py +torchreid/data/datasets/video/prid2011.py +torchreid/engine/__init__.py +torchreid/engine/engine.py +torchreid/engine/image/__init__.py +torchreid/engine/image/softmax.py +torchreid/engine/image/triplet.py +torchreid/engine/video/__init__.py +torchreid/engine/video/softmax.py +torchreid/engine/video/triplet.py +torchreid/losses/__init__.py +torchreid/losses/cross_entropy_loss.py +torchreid/losses/hard_mine_triplet_loss.py +torchreid/metrics/__init__.py +torchreid/metrics/accuracy.py +torchreid/metrics/distance.py +torchreid/metrics/rank.py +torchreid/metrics/rank_cylib/__init__.py +torchreid/metrics/rank_cylib/rank_cy.c +torchreid/metrics/rank_cylib/setup.py +torchreid/metrics/rank_cylib/test_cython.py +torchreid/models/__init__.py +torchreid/models/densenet.py +torchreid/models/hacnn.py +torchreid/models/inceptionresnetv2.py +torchreid/models/inceptionv4.py +torchreid/models/mlfn.py +torchreid/models/mobilenetv2.py +torchreid/models/mudeep.py +torchreid/models/nasnet.py +torchreid/models/osnet.py +torchreid/models/osnet_ain.py +torchreid/models/pcb.py +torchreid/models/resnet.py +torchreid/models/resnet_ibn_a.py +torchreid/models/resnet_ibn_b.py +torchreid/models/resnetmid.py +torchreid/models/senet.py +torchreid/models/shufflenet.py +torchreid/models/shufflenetv2.py +torchreid/models/squeezenet.py +torchreid/models/xception.py +torchreid/optim/__init__.py +torchreid/optim/lr_scheduler.py +torchreid/optim/optimizer.py +torchreid/optim/radam.py +torchreid/utils/__init__.py +torchreid/utils/avgmeter.py +torchreid/utils/feature_extractor.py +torchreid/utils/loggers.py +torchreid/utils/model_complexity.py +torchreid/utils/reidtools.py +torchreid/utils/rerank.py +torchreid/utils/tools.py +torchreid/utils/torchtools.py \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/dependency_links.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/dependency_links.txt new file mode 100644 index 0000000000..8b13789179 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/dependency_links.txt @@ -0,0 +1 @@ + diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/requires.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/requires.txt new file mode 100644 index 0000000000..bb0003a987 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/requires.txt @@ -0,0 +1,16 @@ +numpy +Cython +h5py +Pillow +six +scipy +opencv-python +matplotlib +tb-nightly +future +yacs +gdown +flake8 +yapf +isort +imageio diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/top_level.txt b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/top_level.txt new file mode 100644 index 0000000000..a3cd384f80 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid.egg-info/top_level.txt @@ -0,0 +1 @@ +torchreid -- Gitee From 7fb6c79fd71625f95156a9b563eea75ab997f2ca Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:49:47 +0000 Subject: [PATCH 05/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/test?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/test/env_npu.sh | 69 ------------------- .../OSNet/test/train_eval_8p.sh | 45 ------------ .../OSNet/test/train_finetune_1p.sh | 42 ----------- .../OSNet/test/train_full_1p.sh | 46 ------------- .../OSNet/test/train_full_8p.sh | 45 ------------ .../OSNet/test/train_performance_1p.sh | 46 ------------- .../OSNet/test/train_performance_8p.sh | 45 ------------ 7 files changed, 338 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh delete mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh diff --git a/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh b/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh deleted file mode 100644 index 408016ed19..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh +++ /dev/null @@ -1,69 +0,0 @@ -#!/bin/bash -export install_path=/usr/local/Ascend - -if [ -d ${install_path}/toolkit ]; then - export LD_LIBRARY_PATH=/usr/include/hdf5/lib/:/usr/local/:/usr/local/lib/:/usr/lib/:${install_path}/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons:${path_lib}:${LD_LIBRARY_PATH} - export PATH=${install_path}/fwkacllib/ccec_compiler/bin:${install_path}/fwkacllib/bin:$PATH - export PYTHONPATH=${install_path}/fwkacllib/python/site-packages:${install_path}/tfplugin/python/site-packages:${install_path}/toolkit/python/site-packages:$PYTHONPATH - export PYTHONPATH=/usr/local/python3.7.5/lib/python3.7/site-packages:$PYTHONPATH - export ASCEND_OPP_PATH=${install_path}/opp -else - if [ -d ${install_path}/nnae/latest ];then - export LD_LIBRARY_PATH=/usr/local/:/usr/local/python3.7.5/lib/:/usr/local/openblas/lib:/usr/local/lib/:/usr/lib64/:/usr/lib/:${install_path}/nnae/latest/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons/:/usr/lib/aarch64_64-linux-gnu:$LD_LIBRARY_PATH - export PATH=$PATH:${install_path}/nnae/latest/fwkacllib/ccec_compiler/bin/:${install_path}/nnae/latest/toolkit/tools/ide_daemon/bin/ - export ASCEND_OPP_PATH=${install_path}/nnae/latest/opp/ - export OPTION_EXEC_EXTERN_PLUGIN_PATH=${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libfe.so:${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libaicpu_engine.so:${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libge_local_engine.so - export PYTHONPATH=${install_path}/nnae/latest/fwkacllib/python/site-packages/:${install_path}/nnae/latest/fwkacllib/python/site-packages/auto_tune.egg/auto_tune:${install_path}/nnae/latest/fwkacllib/python/site-packages/schedule_search.egg:$PYTHONPATH - export ASCEND_AICPU_PATH=${install_path}/nnae/latest - else - export LD_LIBRARY_PATH=/usr/local/:/usr/local/lib/:/usr/lib64/:/usr/lib/:/usr/local/python3.7.5/lib/:/usr/local/openblas/lib:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons/:/usr/lib/aarch64-linux-gnu:$LD_LIBRARY_PATH - export PATH=$PATH:${install_path}/ascend-toolkit/latest/fwkacllib/ccec_compiler/bin/:${install_path}/ascend-toolkit/latest/toolkit/tools/ide_daemon/bin/ - export ASCEND_OPP_PATH=${install_path}/ascend-toolkit/latest/opp/ - export OPTION_EXEC_EXTERN_PLUGIN_PATH=${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libfe.so:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libaicpu_engine.so:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libge_local_engine.so - export PYTHONPATH=${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/:${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/auto_tune.egg/auto_tune:${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/schedule_search.egg:$PYTHONPATH - export ASCEND_AICPU_PATH=${install_path}/ascend-toolkit/latest - fi -fi - - -#将Host日志输出到串口,0-关闭/1-开启 -export ASCEND_SLOG_PRINT_TO_STDOUT=0 -#设置默认日志级别,0-debug/1-info/2-warning/3-error -export ASCEND_GLOBAL_LOG_LEVEL=3 -#设置Event日志开启标志,0-关闭/1-开启 -export ASCEND_GLOBAL_EVENT_ENABLE=0 -#设置是否开启taskque,0-关闭/1-开启 -export TASK_QUEUE_ENABLE=1 -#设置是否开启PTCopy,0-关闭/1-开启 -export PTCOPY_ENABLE=1 -#设置是否开启combined标志,0-关闭/1-开启 -export COMBINED_ENABLE=1 -#设置特殊场景是否需要重新编译,不需要修改 -export DYNAMIC_OP="ADD#MUL" -#HCCL白名单开关,1-关闭/0-开启 -export HCCL_WHITELIST_DISABLE=1 -export HCCL_IF_IP=$(hostname -I |awk '{print $1}') - -ulimit -SHn 512000 - -path_lib=$(python3.7 -c """ -import sys -import re -result='' -for index in range(len(sys.path)): - match_sit = re.search('-packages', sys.path[index]) - if match_sit is not None: - match_lib = re.search('lib', sys.path[index]) - - if match_lib is not None: - end=match_lib.span()[1] - result += sys.path[index][0:end] + ':' - - result+=sys.path[index] + '/torch/lib:' -print(result)""" -) - -echo ${path_lib} - -export LD_LIBRARY_PATH=/usr/local/python3.7.5/lib/:${path_lib}:$LD_LIBRARY_PATH - diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh deleted file mode 100644 index a72568ce55..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh +++ /dev/null @@ -1,45 +0,0 @@ -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - echo "[Warning] use default data_path" - data_path="reid-data" - # exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -RANK_ID_START=0 -RANK_SIZE=8 - -for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); -do -KERNEL_NUM=$(($(nproc)/8)) -PID_START=$((KERNEL_NUM * RANK_ID)) -PID_END=$((PID_START + KERNEL_NUM - 1)) -nohup taskset -c $PID_START-$PID_END python3 \ - main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 "test.evaluate" True "model.load_weights" "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" & -done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh deleted file mode 100644 index 3515208263..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh +++ /dev/null @@ -1,42 +0,0 @@ -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - exit 1 -fi - - -weight_info=$2 -weight=`echo ${weight_info#*=}` -if [[ $weight == "" ]];then - echo "[Warning] para \"weight\" not set" - exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -python3 \ - main.py --local_rank 0 --device_num 1 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} --ignore_classifer "train.batch_size" 64 "train.lr" 0.065 "model.load_weights" "${weight}" diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh deleted file mode 100644 index b1d38541e0..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh +++ /dev/null @@ -1,46 +0,0 @@ - -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - echo "[Warning] use default data_path" - data_path="reid-data" - # exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -RANK_ID_START=0 -RANK_SIZE=1 - -for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); -do -KERNEL_NUM=$(($(nproc)/8)) -PID_START=$((KERNEL_NUM * RANK_ID)) -PID_END=$((PID_START + KERNEL_NUM - 1)) -nohup taskset -c $PID_START-$PID_END python3 \ - main.py --local_rank $RANK_ID --device_num ${RANK_SIZE} --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} "train.batch_size" 64 "train.lr" 0.065 & -done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh deleted file mode 100644 index 95347644e0..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh +++ /dev/null @@ -1,45 +0,0 @@ -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - echo "[Warning] use default data_path" - data_path="reid-data" - # exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -RANK_ID_START=0 -RANK_SIZE=8 - -for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); -do -KERNEL_NUM=$(($(nproc)/8)) -PID_START=$((KERNEL_NUM * RANK_ID)) -PID_END=$((PID_START + KERNEL_NUM - 1)) -nohup taskset -c $PID_START-$PID_END python3 \ - main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 & -done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh deleted file mode 100644 index 3f06d07d38..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh +++ /dev/null @@ -1,46 +0,0 @@ - -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - echo "[Warning] use default data_path" - data_path="reid-data" - # exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -RANK_ID_START=0 -RANK_SIZE=1 - -for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); -do -KERNEL_NUM=$(($(nproc)/8)) -PID_START=$((KERNEL_NUM * RANK_ID)) -PID_END=$((PID_START + KERNEL_NUM - 1)) -nohup taskset -c $PID_START-$PID_END python3 \ - main.py --local_rank $RANK_ID --device_num ${RANK_SIZE} --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} "train.batch_size" 64 "train.lr" 0.065 "train.max_epoch" 1 & -done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh deleted file mode 100644 index ffd8fa9eb6..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh +++ /dev/null @@ -1,45 +0,0 @@ -#!/usr/bin/env bash - -data_path_info=$1 -data_path=`echo ${data_path_info#*=}` -if [[ $data_path == "" ]];then - echo "[Warning] para \"data_path\" not set" - echo "[Warning] use default data_path" - data_path="reid-data" - # exit 1 -fi - -###############指定训练脚本执行路径############### -# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 -cur_path=`pwd` -cur_path_last_dirname=${cur_path##*/} -if [ x"${cur_path_last_dirname}" == x"test" ];then - test_path_dir=${cur_path} - cd .. - cur_path=`pwd` -else - test_path_dir=${cur_path}/test -fi - -#################启动训练脚本################# -#训练开始时间，不需要修改 -start_time=$(date +%s) -# 非平台场景时source 环境变量 -check_etp_flag=`env | grep etp_running_flag` -etp_flag=`echo ${check_etp_flag#*=}` -if [ x"${etp_flag}" != x"true" ];then - source ${test_path_dir}/env_npu.sh -fi - -RANK_ID_START=0 -RANK_SIZE=8 - -for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); -do -KERNEL_NUM=$(($(nproc)/8)) -PID_START=$((KERNEL_NUM * RANK_ID)) -PID_END=$((PID_START + KERNEL_NUM - 1)) -nohup taskset -c $PID_START-$PID_END python3 \ - main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ - --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 "train.max_epoch" 1 & -done -- Gitee From 819580cc1df2d8875f9d9c6a745705104e19d374 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:49:54 +0000 Subject: [PATCH 06/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/tools?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../OSNet/tools/compute_mean_std.py | 104 --------- .../OSNet/tools/parse_test_res.py | 148 ------------ .../OSNet/tools/visualize_actmap.py | 219 ------------------ 3 files changed, 471 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py b/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py deleted file mode 100644 index fcedc8173a..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py +++ /dev/null @@ -1,104 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Compute channel-wise mean and standard deviation of a dataset. - -Usage: -$ python compute_mean_std.py DATASET_ROOT DATASET_KEY - -- The first argument points to the root path where you put the datasets. -- The second argument means the specific dataset key. - -For instance, your datasets are put under $DATA and you wanna -compute the statistics of Market1501, do -$ python compute_mean_std.py $DATA market1501 -""" -import argparse - -import torchreid - - -def main(): - parser = argparse.ArgumentParser() - parser.add_argument('root', type=str) - parser.add_argument('sources', type=str) - args = parser.parse_args() - - datamanager = torchreid.data.ImageDataManager( - root=args.root, - sources=args.sources, - targets=None, - height=256, - width=128, - batch_size_train=100, - batch_size_test=100, - transforms=None, - norm_mean=[0., 0., 0.], - norm_std=[1., 1., 1.], - train_sampler='SequentialSampler' - ) - train_loader = datamanager.train_loader - - print('Computing mean and std ...') - mean = 0. - std = 0. - n_samples = 0. - for data in train_loader: - data = data['img'] - batch_size = data.size(0) - data = data.view(batch_size, data.size(1), -1) - mean += data.mean(2).sum(0) - std += data.std(2).sum(0) - n_samples += batch_size - - mean /= n_samples - std /= n_samples - print('Mean: {}'.format(mean)) - print('Std: {}'.format(std)) - - -if __name__ == '__main__': - main() diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py b/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py deleted file mode 100644 index bbd143e4ff..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py +++ /dev/null @@ -1,148 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -This script aims to automate the process of calculating average results -stored in the test.log files over multiple splits. - -How to use: -For example, you have done evaluation over 20 splits on VIPeR, leading to -the following file structure - -log/ - eval_viper/ - split_0/ - test.log-xxxx - split_1/ - test.log-xxxx - split_2/ - test.log-xxxx - ... - -You can run the following command in your terminal to get the average performance: -$ python tools/parse_test_res.py log/eval_viper -""" -import os -import re -import glob -import numpy as np -import argparse -from collections import defaultdict - -from torchreid.utils import check_isfile, listdir_nohidden - - -def parse_file(filepath, regex_mAP, regex_r1, regex_r5, regex_r10, regex_r20): - results = {} - - with open(filepath, 'r') as f: - lines = f.readlines() - - for line in lines: - line = line.strip() - - match_mAP = regex_mAP.search(line) - if match_mAP: - mAP = float(match_mAP.group(1)) - results['mAP'] = mAP - - match_r1 = regex_r1.search(line) - if match_r1: - r1 = float(match_r1.group(1)) - results['r1'] = r1 - - match_r5 = regex_r5.search(line) - if match_r5: - r5 = float(match_r5.group(1)) - results['r5'] = r5 - - match_r10 = regex_r10.search(line) - if match_r10: - r10 = float(match_r10.group(1)) - results['r10'] = r10 - - match_r20 = regex_r20.search(line) - if match_r20: - r20 = float(match_r20.group(1)) - results['r20'] = r20 - - return results - - -def main(args): - regex_mAP = re.compile(r'mAP: ([\.\deE+-]+)%') - regex_r1 = re.compile(r'Rank-1 : ([\.\deE+-]+)%') - regex_r5 = re.compile(r'Rank-5 : ([\.\deE+-]+)%') - regex_r10 = re.compile(r'Rank-10 : ([\.\deE+-]+)%') - regex_r20 = re.compile(r'Rank-20 : ([\.\deE+-]+)%') - - final_res = defaultdict(list) - - directories = listdir_nohidden(args.directory, sort=True) - num_dirs = len(directories) - for directory in directories: - fullpath = os.path.join(args.directory, directory) - filepath = glob.glob(os.path.join(fullpath, 'test.log*'))[0] - check_isfile(filepath) - print(f'Parsing {filepath}') - res = parse_file( - filepath, regex_mAP, regex_r1, regex_r5, regex_r10, regex_r20 - ) - for key, value in res.items(): - final_res[key].append(value) - - print('Finished parsing') - print(f'The average results over {num_dirs} splits are shown below') - - for key, values in final_res.items(): - mean_val = np.mean(values) - print(f'{key}: {mean_val:.1f}') - - -if __name__ == '__main__': - parser = argparse.ArgumentParser() - parser.add_argument('directory', type=str, help='Path to directory') - args = parser.parse_args() - main(args) diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py b/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py deleted file mode 100644 index e9d6586d8e..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py +++ /dev/null @@ -1,219 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Visualizes CNN activation maps to see where the CNN focuses on to extract features. - -Reference: - - Zagoruyko and Komodakis. Paying more attention to attention: Improving the - performance of convolutional neural networks via attention transfer. ICLR, 2017 - - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. -""" -import numpy as np -import os.path as osp -import argparse -import cv2 -import torch -from torch.nn import functional as F - -import torchreid -from torchreid.utils import ( - check_isfile, mkdir_if_missing, load_pretrained_weights -) - -IMAGENET_MEAN = [0.485, 0.456, 0.406] -IMAGENET_STD = [0.229, 0.224, 0.225] -GRID_SPACING = 10 - - -@torch.no_grad() -def visactmap( - model, - test_loader, - save_dir, - width, - height, - use_gpu, - img_mean=None, - img_std=None -): - if img_mean is None or img_std is None: - # use imagenet mean and std - img_mean = IMAGENET_MEAN - img_std = IMAGENET_STD - - model.eval() - - for target in list(test_loader.keys()): - data_loader = test_loader[target]['query'] # only process query images - # original images and activation maps are saved individually - actmap_dir = osp.join(save_dir, 'actmap_' + target) - mkdir_if_missing(actmap_dir) - print('Visualizing activation maps for {} ...'.format(target)) - - for batch_idx, data in enumerate(data_loader): - imgs, paths = data['img'], data['impath'] - if use_gpu: - imgs = imgs.cuda() - - # forward to get convolutional feature maps - try: - outputs = model(imgs, return_featuremaps=True) - except TypeError: - raise TypeError( - 'forward() got unexpected keyword argument "return_featuremaps". ' - 'Please add return_featuremaps as an input argument to forward(). When ' - 'return_featuremaps=True, return feature maps only.' - ) - - if outputs.dim() != 4: - raise ValueError( - 'The model output is supposed to have ' - 'shape of (b, c, h, w), i.e. 4 dimensions, but got {} dimensions. ' - 'Please make sure you set the model output at eval mode ' - 'to be the last convolutional feature maps'.format( - outputs.dim() - ) - ) - - # compute activation maps - outputs = (outputs**2).sum(1) - b, h, w = outputs.size() - outputs = outputs.view(b, h * w) - outputs = F.normalize(outputs, p=2, dim=1) - outputs = outputs.view(b, h, w) - - if use_gpu: - imgs, outputs = imgs.cpu(), outputs.cpu() - - for j in range(outputs.size(0)): - # get image name - path = paths[j] - imname = osp.basename(osp.splitext(path)[0]) - - # RGB image - img = imgs[j, ...] - for t, m, s in zip(img, img_mean, img_std): - t.mul_(s).add_(m).clamp_(0, 1) - img_np = np.uint8(np.floor(img.numpy() * 255)) - img_np = img_np.transpose((1, 2, 0)) # (c, h, w) -> (h, w, c) - - # activation map - am = outputs[j, ...].numpy() - am = cv2.resize(am, (width, height)) - am = 255 * (am - np.min(am)) / ( - np.max(am) - np.min(am) + 1e-12 - ) - am = np.uint8(np.floor(am)) - am = cv2.applyColorMap(am, cv2.COLORMAP_JET) - - # overlapped - overlapped = img_np*0.3 + am*0.7 - overlapped[overlapped > 255] = 255 - overlapped = overlapped.astype(np.uint8) - - # save images in a single figure (add white spacing between images) - # from left to right: original image, activation map, overlapped image - grid_img = 255 * np.ones( - (height, 3*width + 2*GRID_SPACING, 3), dtype=np.uint8 - ) - grid_img[:, :width, :] = img_np[:, :, ::-1] - grid_img[:, - width + GRID_SPACING:2*width + GRID_SPACING, :] = am - grid_img[:, 2*width + 2*GRID_SPACING:, :] = overlapped - cv2.imwrite(osp.join(actmap_dir, imname + '.jpg'), grid_img) - - if (batch_idx+1) % 10 == 0: - print( - '- done batch {}/{}'.format( - batch_idx + 1, len(data_loader) - ) - ) - - -def main(): - parser = argparse.ArgumentParser() - parser.add_argument('--root', type=str) - parser.add_argument('-d', '--dataset', type=str, default='market1501') - parser.add_argument('-m', '--model', type=str, default='osnet_x1_0') - parser.add_argument('--weights', type=str) - parser.add_argument('--save-dir', type=str, default='log') - parser.add_argument('--height', type=int, default=256) - parser.add_argument('--width', type=int, default=128) - args = parser.parse_args() - - use_gpu = torch.cuda.is_available() - - datamanager = torchreid.data.ImageDataManager( - root=args.root, - sources=args.dataset, - height=args.height, - width=args.width, - batch_size_train=100, - batch_size_test=100, - transforms=None, - train_sampler='SequentialSampler' - ) - test_loader = datamanager.test_loader - - model = torchreid.models.build_model( - name=args.model, - num_classes=datamanager.num_train_pids, - use_gpu=use_gpu - ) - - if use_gpu: - model = model.cuda() - - if args.weights and check_isfile(args.weights): - load_pretrained_weights(model, args.weights) - - visactmap( - model, test_loader, args.save_dir, args.width, args.height, use_gpu - ) - - -if __name__ == '__main__': - main() -- Gitee From 25d0df66113011e13ebd48b6114cd68b11ee49f2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:50:08 +0000 Subject: [PATCH 07/31] init --- .../cv/classification/OSNet/test/env_npu.sh | 69 +++++++++++++++++++ .../OSNet/test/train_eval_8p.sh | 45 ++++++++++++ .../OSNet/test/train_finetune_1p.sh | 42 +++++++++++ .../OSNet/test/train_full_1p.sh | 46 +++++++++++++ .../OSNet/test/train_full_8p.sh | 45 ++++++++++++ .../OSNet/test/train_performance_1p.sh | 46 +++++++++++++ .../OSNet/test/train_performance_8p.sh | 45 ++++++++++++ 7 files changed, 338 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh create mode 100644 PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh diff --git a/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh b/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh new file mode 100644 index 0000000000..408016ed19 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/env_npu.sh @@ -0,0 +1,69 @@ +#!/bin/bash +export install_path=/usr/local/Ascend + +if [ -d ${install_path}/toolkit ]; then + export LD_LIBRARY_PATH=/usr/include/hdf5/lib/:/usr/local/:/usr/local/lib/:/usr/lib/:${install_path}/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons:${path_lib}:${LD_LIBRARY_PATH} + export PATH=${install_path}/fwkacllib/ccec_compiler/bin:${install_path}/fwkacllib/bin:$PATH + export PYTHONPATH=${install_path}/fwkacllib/python/site-packages:${install_path}/tfplugin/python/site-packages:${install_path}/toolkit/python/site-packages:$PYTHONPATH + export PYTHONPATH=/usr/local/python3.7.5/lib/python3.7/site-packages:$PYTHONPATH + export ASCEND_OPP_PATH=${install_path}/opp +else + if [ -d ${install_path}/nnae/latest ];then + export LD_LIBRARY_PATH=/usr/local/:/usr/local/python3.7.5/lib/:/usr/local/openblas/lib:/usr/local/lib/:/usr/lib64/:/usr/lib/:${install_path}/nnae/latest/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons/:/usr/lib/aarch64_64-linux-gnu:$LD_LIBRARY_PATH + export PATH=$PATH:${install_path}/nnae/latest/fwkacllib/ccec_compiler/bin/:${install_path}/nnae/latest/toolkit/tools/ide_daemon/bin/ + export ASCEND_OPP_PATH=${install_path}/nnae/latest/opp/ + export OPTION_EXEC_EXTERN_PLUGIN_PATH=${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libfe.so:${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libaicpu_engine.so:${install_path}/nnae/latest/fwkacllib/lib64/plugin/opskernel/libge_local_engine.so + export PYTHONPATH=${install_path}/nnae/latest/fwkacllib/python/site-packages/:${install_path}/nnae/latest/fwkacllib/python/site-packages/auto_tune.egg/auto_tune:${install_path}/nnae/latest/fwkacllib/python/site-packages/schedule_search.egg:$PYTHONPATH + export ASCEND_AICPU_PATH=${install_path}/nnae/latest + else + export LD_LIBRARY_PATH=/usr/local/:/usr/local/lib/:/usr/lib64/:/usr/lib/:/usr/local/python3.7.5/lib/:/usr/local/openblas/lib:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/:${install_path}/driver/lib64/common/:${install_path}/driver/lib64/driver/:${install_path}/add-ons/:/usr/lib/aarch64-linux-gnu:$LD_LIBRARY_PATH + export PATH=$PATH:${install_path}/ascend-toolkit/latest/fwkacllib/ccec_compiler/bin/:${install_path}/ascend-toolkit/latest/toolkit/tools/ide_daemon/bin/ + export ASCEND_OPP_PATH=${install_path}/ascend-toolkit/latest/opp/ + export OPTION_EXEC_EXTERN_PLUGIN_PATH=${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libfe.so:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libaicpu_engine.so:${install_path}/ascend-toolkit/latest/fwkacllib/lib64/plugin/opskernel/libge_local_engine.so + export PYTHONPATH=${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/:${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/auto_tune.egg/auto_tune:${install_path}/ascend-toolkit/latest/fwkacllib/python/site-packages/schedule_search.egg:$PYTHONPATH + export ASCEND_AICPU_PATH=${install_path}/ascend-toolkit/latest + fi +fi + + +#将Host日志输出到串口,0-关闭/1-开启 +export ASCEND_SLOG_PRINT_TO_STDOUT=0 +#设置默认日志级别,0-debug/1-info/2-warning/3-error +export ASCEND_GLOBAL_LOG_LEVEL=3 +#设置Event日志开启标志,0-关闭/1-开启 +export ASCEND_GLOBAL_EVENT_ENABLE=0 +#设置是否开启taskque,0-关闭/1-开启 +export TASK_QUEUE_ENABLE=1 +#设置是否开启PTCopy,0-关闭/1-开启 +export PTCOPY_ENABLE=1 +#设置是否开启combined标志,0-关闭/1-开启 +export COMBINED_ENABLE=1 +#设置特殊场景是否需要重新编译,不需要修改 +export DYNAMIC_OP="ADD#MUL" +#HCCL白名单开关,1-关闭/0-开启 +export HCCL_WHITELIST_DISABLE=1 +export HCCL_IF_IP=$(hostname -I |awk '{print $1}') + +ulimit -SHn 512000 + +path_lib=$(python3.7 -c """ +import sys +import re +result='' +for index in range(len(sys.path)): + match_sit = re.search('-packages', sys.path[index]) + if match_sit is not None: + match_lib = re.search('lib', sys.path[index]) + + if match_lib is not None: + end=match_lib.span()[1] + result += sys.path[index][0:end] + ':' + + result+=sys.path[index] + '/torch/lib:' +print(result)""" +) + +echo ${path_lib} + +export LD_LIBRARY_PATH=/usr/local/python3.7.5/lib/:${path_lib}:$LD_LIBRARY_PATH + diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh new file mode 100644 index 0000000000..a72568ce55 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_eval_8p.sh @@ -0,0 +1,45 @@ +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + echo "[Warning] use default data_path" + data_path="reid-data" + # exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +RANK_ID_START=0 +RANK_SIZE=8 + +for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); +do +KERNEL_NUM=$(($(nproc)/8)) +PID_START=$((KERNEL_NUM * RANK_ID)) +PID_END=$((PID_START + KERNEL_NUM - 1)) +nohup taskset -c $PID_START-$PID_END python3 \ + main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 "test.evaluate" True "model.load_weights" "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" & +done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh new file mode 100644 index 0000000000..3515208263 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_finetune_1p.sh @@ -0,0 +1,42 @@ +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + exit 1 +fi + + +weight_info=$2 +weight=`echo ${weight_info#*=}` +if [[ $weight == "" ]];then + echo "[Warning] para \"weight\" not set" + exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +python3 \ + main.py --local_rank 0 --device_num 1 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} --ignore_classifer "train.batch_size" 64 "train.lr" 0.065 "model.load_weights" "${weight}" diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh new file mode 100644 index 0000000000..b1d38541e0 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_full_1p.sh @@ -0,0 +1,46 @@ + +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + echo "[Warning] use default data_path" + data_path="reid-data" + # exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +RANK_ID_START=0 +RANK_SIZE=1 + +for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); +do +KERNEL_NUM=$(($(nproc)/8)) +PID_START=$((KERNEL_NUM * RANK_ID)) +PID_END=$((PID_START + KERNEL_NUM - 1)) +nohup taskset -c $PID_START-$PID_END python3 \ + main.py --local_rank $RANK_ID --device_num ${RANK_SIZE} --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} "train.batch_size" 64 "train.lr" 0.065 & +done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh new file mode 100644 index 0000000000..95347644e0 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_full_8p.sh @@ -0,0 +1,45 @@ +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + echo "[Warning] use default data_path" + data_path="reid-data" + # exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +RANK_ID_START=0 +RANK_SIZE=8 + +for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); +do +KERNEL_NUM=$(($(nproc)/8)) +PID_START=$((KERNEL_NUM * RANK_ID)) +PID_END=$((PID_START + KERNEL_NUM - 1)) +nohup taskset -c $PID_START-$PID_END python3 \ + main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 & +done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh new file mode 100644 index 0000000000..3f06d07d38 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_1p.sh @@ -0,0 +1,46 @@ + +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + echo "[Warning] use default data_path" + data_path="reid-data" + # exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +RANK_ID_START=0 +RANK_SIZE=1 + +for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); +do +KERNEL_NUM=$(($(nproc)/8)) +PID_START=$((KERNEL_NUM * RANK_ID)) +PID_END=$((PID_START + KERNEL_NUM - 1)) +nohup taskset -c $PID_START-$PID_END python3 \ + main.py --local_rank $RANK_ID --device_num ${RANK_SIZE} --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} "train.batch_size" 64 "train.lr" 0.065 "train.max_epoch" 1 & +done diff --git a/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh new file mode 100644 index 0000000000..ffd8fa9eb6 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/test/train_performance_8p.sh @@ -0,0 +1,45 @@ +#!/usr/bin/env bash + +data_path_info=$1 +data_path=`echo ${data_path_info#*=}` +if [[ $data_path == "" ]];then + echo "[Warning] para \"data_path\" not set" + echo "[Warning] use default data_path" + data_path="reid-data" + # exit 1 +fi + +###############指定训练脚本执行路径############### +# cd到与test文件夹同层级目录下执行脚本，提高兼容性；test_path_dir为包含test文件夹的路径 +cur_path=`pwd` +cur_path_last_dirname=${cur_path##*/} +if [ x"${cur_path_last_dirname}" == x"test" ];then + test_path_dir=${cur_path} + cd .. + cur_path=`pwd` +else + test_path_dir=${cur_path}/test +fi + +#################启动训练脚本################# +#训练开始时间，不需要修改 +start_time=$(date +%s) +# 非平台场景时source 环境变量 +check_etp_flag=`env | grep etp_running_flag` +etp_flag=`echo ${check_etp_flag#*=}` +if [ x"${etp_flag}" != x"true" ];then + source ${test_path_dir}/env_npu.sh +fi + +RANK_ID_START=0 +RANK_SIZE=8 + +for((RANK_ID=$RANK_ID_START;RANK_ID<$((RANK_SIZE+RANK_ID_START));RANK_ID++)); +do +KERNEL_NUM=$(($(nproc)/8)) +PID_START=$((KERNEL_NUM * RANK_ID)) +PID_END=$((PID_START + KERNEL_NUM - 1)) +nohup taskset -c $PID_START-$PID_END python3 \ + main.py --local_rank $RANK_ID --device_num 8 --config-file "configs/osnet_x1_0_trained_from_scratch.yaml" --npu --amp \ + --root ${data_path} "train.batch_size" 32 "train.lr" 0.26 "train.max_epoch" 1 & +done -- Gitee From 43012217bb15050b5bb9974e524a033f4cdb59a2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:50:22 +0000 Subject: [PATCH 08/31] init --- .../OSNet/tools/compute_mean_std.py | 104 +++++++++ .../OSNet/tools/parse_test_res.py | 148 ++++++++++++ .../OSNet/tools/visualize_actmap.py | 219 ++++++++++++++++++ 3 files changed, 471 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py b/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py new file mode 100644 index 0000000000..fcedc8173a --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/tools/compute_mean_std.py @@ -0,0 +1,104 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Compute channel-wise mean and standard deviation of a dataset. + +Usage: +$ python compute_mean_std.py DATASET_ROOT DATASET_KEY + +- The first argument points to the root path where you put the datasets. +- The second argument means the specific dataset key. + +For instance, your datasets are put under $DATA and you wanna +compute the statistics of Market1501, do +$ python compute_mean_std.py $DATA market1501 +""" +import argparse + +import torchreid + + +def main(): + parser = argparse.ArgumentParser() + parser.add_argument('root', type=str) + parser.add_argument('sources', type=str) + args = parser.parse_args() + + datamanager = torchreid.data.ImageDataManager( + root=args.root, + sources=args.sources, + targets=None, + height=256, + width=128, + batch_size_train=100, + batch_size_test=100, + transforms=None, + norm_mean=[0., 0., 0.], + norm_std=[1., 1., 1.], + train_sampler='SequentialSampler' + ) + train_loader = datamanager.train_loader + + print('Computing mean and std ...') + mean = 0. + std = 0. + n_samples = 0. + for data in train_loader: + data = data['img'] + batch_size = data.size(0) + data = data.view(batch_size, data.size(1), -1) + mean += data.mean(2).sum(0) + std += data.std(2).sum(0) + n_samples += batch_size + + mean /= n_samples + std /= n_samples + print('Mean: {}'.format(mean)) + print('Std: {}'.format(std)) + + +if __name__ == '__main__': + main() diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py b/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py new file mode 100644 index 0000000000..bbd143e4ff --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/tools/parse_test_res.py @@ -0,0 +1,148 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +This script aims to automate the process of calculating average results +stored in the test.log files over multiple splits. + +How to use: +For example, you have done evaluation over 20 splits on VIPeR, leading to +the following file structure + +log/ + eval_viper/ + split_0/ + test.log-xxxx + split_1/ + test.log-xxxx + split_2/ + test.log-xxxx + ... + +You can run the following command in your terminal to get the average performance: +$ python tools/parse_test_res.py log/eval_viper +""" +import os +import re +import glob +import numpy as np +import argparse +from collections import defaultdict + +from torchreid.utils import check_isfile, listdir_nohidden + + +def parse_file(filepath, regex_mAP, regex_r1, regex_r5, regex_r10, regex_r20): + results = {} + + with open(filepath, 'r') as f: + lines = f.readlines() + + for line in lines: + line = line.strip() + + match_mAP = regex_mAP.search(line) + if match_mAP: + mAP = float(match_mAP.group(1)) + results['mAP'] = mAP + + match_r1 = regex_r1.search(line) + if match_r1: + r1 = float(match_r1.group(1)) + results['r1'] = r1 + + match_r5 = regex_r5.search(line) + if match_r5: + r5 = float(match_r5.group(1)) + results['r5'] = r5 + + match_r10 = regex_r10.search(line) + if match_r10: + r10 = float(match_r10.group(1)) + results['r10'] = r10 + + match_r20 = regex_r20.search(line) + if match_r20: + r20 = float(match_r20.group(1)) + results['r20'] = r20 + + return results + + +def main(args): + regex_mAP = re.compile(r'mAP: ([\.\deE+-]+)%') + regex_r1 = re.compile(r'Rank-1 : ([\.\deE+-]+)%') + regex_r5 = re.compile(r'Rank-5 : ([\.\deE+-]+)%') + regex_r10 = re.compile(r'Rank-10 : ([\.\deE+-]+)%') + regex_r20 = re.compile(r'Rank-20 : ([\.\deE+-]+)%') + + final_res = defaultdict(list) + + directories = listdir_nohidden(args.directory, sort=True) + num_dirs = len(directories) + for directory in directories: + fullpath = os.path.join(args.directory, directory) + filepath = glob.glob(os.path.join(fullpath, 'test.log*'))[0] + check_isfile(filepath) + print(f'Parsing {filepath}') + res = parse_file( + filepath, regex_mAP, regex_r1, regex_r5, regex_r10, regex_r20 + ) + for key, value in res.items(): + final_res[key].append(value) + + print('Finished parsing') + print(f'The average results over {num_dirs} splits are shown below') + + for key, values in final_res.items(): + mean_val = np.mean(values) + print(f'{key}: {mean_val:.1f}') + + +if __name__ == '__main__': + parser = argparse.ArgumentParser() + parser.add_argument('directory', type=str, help='Path to directory') + args = parser.parse_args() + main(args) diff --git a/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py b/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py new file mode 100644 index 0000000000..e9d6586d8e --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/tools/visualize_actmap.py @@ -0,0 +1,219 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Visualizes CNN activation maps to see where the CNN focuses on to extract features. + +Reference: + - Zagoruyko and Komodakis. Paying more attention to attention: Improving the + performance of convolutional neural networks via attention transfer. ICLR, 2017 + - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. +""" +import numpy as np +import os.path as osp +import argparse +import cv2 +import torch +from torch.nn import functional as F + +import torchreid +from torchreid.utils import ( + check_isfile, mkdir_if_missing, load_pretrained_weights +) + +IMAGENET_MEAN = [0.485, 0.456, 0.406] +IMAGENET_STD = [0.229, 0.224, 0.225] +GRID_SPACING = 10 + + +@torch.no_grad() +def visactmap( + model, + test_loader, + save_dir, + width, + height, + use_gpu, + img_mean=None, + img_std=None +): + if img_mean is None or img_std is None: + # use imagenet mean and std + img_mean = IMAGENET_MEAN + img_std = IMAGENET_STD + + model.eval() + + for target in list(test_loader.keys()): + data_loader = test_loader[target]['query'] # only process query images + # original images and activation maps are saved individually + actmap_dir = osp.join(save_dir, 'actmap_' + target) + mkdir_if_missing(actmap_dir) + print('Visualizing activation maps for {} ...'.format(target)) + + for batch_idx, data in enumerate(data_loader): + imgs, paths = data['img'], data['impath'] + if use_gpu: + imgs = imgs.cuda() + + # forward to get convolutional feature maps + try: + outputs = model(imgs, return_featuremaps=True) + except TypeError: + raise TypeError( + 'forward() got unexpected keyword argument "return_featuremaps". ' + 'Please add return_featuremaps as an input argument to forward(). When ' + 'return_featuremaps=True, return feature maps only.' + ) + + if outputs.dim() != 4: + raise ValueError( + 'The model output is supposed to have ' + 'shape of (b, c, h, w), i.e. 4 dimensions, but got {} dimensions. ' + 'Please make sure you set the model output at eval mode ' + 'to be the last convolutional feature maps'.format( + outputs.dim() + ) + ) + + # compute activation maps + outputs = (outputs**2).sum(1) + b, h, w = outputs.size() + outputs = outputs.view(b, h * w) + outputs = F.normalize(outputs, p=2, dim=1) + outputs = outputs.view(b, h, w) + + if use_gpu: + imgs, outputs = imgs.cpu(), outputs.cpu() + + for j in range(outputs.size(0)): + # get image name + path = paths[j] + imname = osp.basename(osp.splitext(path)[0]) + + # RGB image + img = imgs[j, ...] + for t, m, s in zip(img, img_mean, img_std): + t.mul_(s).add_(m).clamp_(0, 1) + img_np = np.uint8(np.floor(img.numpy() * 255)) + img_np = img_np.transpose((1, 2, 0)) # (c, h, w) -> (h, w, c) + + # activation map + am = outputs[j, ...].numpy() + am = cv2.resize(am, (width, height)) + am = 255 * (am - np.min(am)) / ( + np.max(am) - np.min(am) + 1e-12 + ) + am = np.uint8(np.floor(am)) + am = cv2.applyColorMap(am, cv2.COLORMAP_JET) + + # overlapped + overlapped = img_np*0.3 + am*0.7 + overlapped[overlapped > 255] = 255 + overlapped = overlapped.astype(np.uint8) + + # save images in a single figure (add white spacing between images) + # from left to right: original image, activation map, overlapped image + grid_img = 255 * np.ones( + (height, 3*width + 2*GRID_SPACING, 3), dtype=np.uint8 + ) + grid_img[:, :width, :] = img_np[:, :, ::-1] + grid_img[:, + width + GRID_SPACING:2*width + GRID_SPACING, :] = am + grid_img[:, 2*width + 2*GRID_SPACING:, :] = overlapped + cv2.imwrite(osp.join(actmap_dir, imname + '.jpg'), grid_img) + + if (batch_idx+1) % 10 == 0: + print( + '- done batch {}/{}'.format( + batch_idx + 1, len(data_loader) + ) + ) + + +def main(): + parser = argparse.ArgumentParser() + parser.add_argument('--root', type=str) + parser.add_argument('-d', '--dataset', type=str, default='market1501') + parser.add_argument('-m', '--model', type=str, default='osnet_x1_0') + parser.add_argument('--weights', type=str) + parser.add_argument('--save-dir', type=str, default='log') + parser.add_argument('--height', type=int, default=256) + parser.add_argument('--width', type=int, default=128) + args = parser.parse_args() + + use_gpu = torch.cuda.is_available() + + datamanager = torchreid.data.ImageDataManager( + root=args.root, + sources=args.dataset, + height=args.height, + width=args.width, + batch_size_train=100, + batch_size_test=100, + transforms=None, + train_sampler='SequentialSampler' + ) + test_loader = datamanager.test_loader + + model = torchreid.models.build_model( + name=args.model, + num_classes=datamanager.num_train_pids, + use_gpu=use_gpu + ) + + if use_gpu: + model = model.cuda() + + if args.weights and check_isfile(args.weights): + load_pretrained_weights(model, args.weights) + + visactmap( + model, test_loader, args.save_dir, args.width, args.height, use_gpu + ) + + +if __name__ == '__main__': + main() -- Gitee From fb2effc5449b4aefb57b053a773f0ee4c262962b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:50:38 +0000 Subject: [PATCH 09/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/torchreid?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../OSNet/torchreid/__init__.py | 56 - .../OSNet/torchreid/data/__init__.py | 54 - .../OSNet/torchreid/data/datamanager.py | 609 --------- .../OSNet/torchreid/data/datasets/__init__.py | 166 --- .../OSNet/torchreid/data/datasets/dataset.py | 529 -------- .../torchreid/data/datasets/image/__init__.py | 62 - .../torchreid/data/datasets/image/cuhk01.py | 184 --- .../torchreid/data/datasets/image/cuhk02.py | 144 -- .../torchreid/data/datasets/image/cuhk03.py | 354 ----- .../torchreid/data/datasets/image/cuhksysu.py | 107 -- .../data/datasets/image/dukemtmcreid.py | 115 -- .../torchreid/data/datasets/image/grid.py | 178 --- .../torchreid/data/datasets/image/ilids.py | 182 --- .../data/datasets/image/market1501.py | 133 -- .../torchreid/data/datasets/image/msmt17.py | 145 -- .../torchreid/data/datasets/image/prid.py | 154 --- .../data/datasets/image/sensereid.py | 117 -- .../data/datasets/image/university1652.py | 157 --- .../torchreid/data/datasets/image/viper.py | 175 --- .../torchreid/data/datasets/video/__init__.py | 53 - .../data/datasets/video/dukemtmcvidreid.py | 175 --- .../torchreid/data/datasets/video/ilidsvid.py | 190 --- .../torchreid/data/datasets/video/mars.py | 180 --- .../torchreid/data/datasets/video/prid2011.py | 127 -- .../OSNet/torchreid/data/sampler.py | 292 ---- .../OSNet/torchreid/data/transforms.py | 373 ------ .../OSNet/torchreid/engine/__init__.py | 52 - .../OSNet/torchreid/engine/engine.py | 547 -------- .../OSNet/torchreid/engine/image/__init__.py | 51 - .../OSNet/torchreid/engine/image/softmax.py | 157 --- .../OSNet/torchreid/engine/image/triplet.py | 169 --- .../OSNet/torchreid/engine/video/__init__.py | 51 - .../OSNet/torchreid/engine/video/softmax.py | 156 --- .../OSNet/torchreid/engine/video/triplet.py | 169 --- .../OSNet/torchreid/losses/__init__.py | 68 - .../torchreid/losses/cross_entropy_loss.py | 100 -- .../losses/hard_mine_triplet_loss.py | 95 -- .../OSNet/torchreid/metrics/__init__.py | 52 - .../OSNet/torchreid/metrics/accuracy.py | 84 -- .../OSNet/torchreid/metrics/distance.py | 127 -- .../OSNet/torchreid/metrics/rank.py | 254 ---- .../torchreid/metrics/rank_cylib/Makefile | 6 - .../torchreid/metrics/rank_cylib/__init__.py | 47 - .../torchreid/metrics/rank_cylib/rank_cy.pyx | 251 ---- .../torchreid/metrics/rank_cylib/setup.py | 73 - .../metrics/rank_cylib/test_cython.py | 130 -- .../OSNet/torchreid/models/__init__.py | 166 --- .../OSNet/torchreid/models/densenet.py | 425 ------ .../OSNet/torchreid/models/hacnn.py | 461 ------- .../torchreid/models/inceptionresnetv2.py | 406 ------ .../OSNet/torchreid/models/inceptionv4.py | 428 ------ .../OSNet/torchreid/models/mlfn.py | 316 ----- .../OSNet/torchreid/models/mobilenetv2.py | 321 ----- .../OSNet/torchreid/models/mudeep.py | 253 ---- .../OSNet/torchreid/models/nasnet.py | 1178 ----------------- .../OSNet/torchreid/models/osnet.py | 645 --------- .../OSNet/torchreid/models/osnet_ain.py | 588 -------- .../OSNet/torchreid/models/pcb.py | 361 ----- .../OSNet/torchreid/models/resnet.py | 576 -------- .../OSNet/torchreid/models/resnet_ibn_a.py | 334 ----- .../OSNet/torchreid/models/resnet_ibn_b.py | 319 ----- .../OSNet/torchreid/models/resnetmid.py | 354 ----- .../OSNet/torchreid/models/senet.py | 735 ---------- .../OSNet/torchreid/models/shufflenet.py | 245 ---- .../OSNet/torchreid/models/shufflenetv2.py | 307 ----- .../OSNet/torchreid/models/squeezenet.py | 281 ---- .../OSNet/torchreid/models/xception.py | 391 ------ .../OSNet/torchreid/optim/__init__.py | 51 - .../OSNet/torchreid/optim/lr_scheduler.py | 115 -- .../OSNet/torchreid/optim/optimizer.py | 218 --- .../OSNet/torchreid/optim/radam.py | 376 ------ .../OSNet/torchreid/utils/__init__.py | 57 - .../OSNet/torchreid/utils/avgmeter.py | 120 -- .../torchreid/utils/feature_extractor.py | 199 --- .../OSNet/torchreid/utils/loggers.py | 193 --- .../OSNet/torchreid/utils/model_complexity.py | 410 ------ .../OSNet/torchreid/utils/reidtools.py | 201 --- .../OSNet/torchreid/utils/rerank.py | 157 --- .../OSNet/torchreid/utils/tools.py | 189 --- .../OSNet/torchreid/utils/torchtools.py | 363 ----- 80 files changed, 19389 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py deleted file mode 100644 index 40669b944c..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py +++ /dev/null @@ -1,56 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from torchreid import data, optim, utils, engine, losses, models, metrics - -__version__ = '1.4.0' -__author__ = 'Kaiyang Zhou' -__homepage__ = 'https://kaiyangzhou.github.io/' -__description__ = 'Deep learning person re-identification in PyTorch' -__url__ = 'https://github.com/KaiyangZhou/deep-person-reid' diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py deleted file mode 100644 index 577ed45750..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py +++ /dev/null @@ -1,54 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .datasets import ( - Dataset, ImageDataset, VideoDataset, register_image_dataset, - register_video_dataset -) -from .datamanager import ImageDataManager, VideoDataManager diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py deleted file mode 100644 index 009d03802f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py +++ /dev/null @@ -1,609 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch - -from torchreid.data.sampler import build_train_sampler -from torchreid.data.datasets import init_image_dataset, init_video_dataset -from torchreid.data.transforms import build_transforms - - -class DataManager(object): - r"""Base data manager. - - Args: - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - """ - - def __init__( - self, - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - norm_mean=None, - norm_std=None, - use_gpu=False - ): - self.sources = sources - self.targets = targets - self.height = height - self.width = width - - if self.sources is None: - raise ValueError('sources must not be None') - - if isinstance(self.sources, str): - self.sources = [self.sources] - - if self.targets is None: - self.targets = self.sources - - if isinstance(self.targets, str): - self.targets = [self.targets] - - self.transform_tr, self.transform_te = build_transforms( - self.height, - self.width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std - ) - - self.use_gpu = (torch.cuda.is_available() and use_gpu) - - @property - def num_train_pids(self): - """Returns the number of training person identities.""" - return self._num_train_pids - - @property - def num_train_cams(self): - """Returns the number of training cameras.""" - return self._num_train_cams - - def fetch_test_loaders(self, name): - """Returns query and gallery of a test dataset, each containing - tuples of (img_path(s), pid, camid). - - Args: - name (str): dataset name. - """ - query_loader = self.test_dataset[name]['query'] - gallery_loader = self.test_dataset[name]['gallery'] - return query_loader, gallery_loader - - def preprocess_pil_img(self, img): - """Transforms a PIL image to torch tensor for testing.""" - return self.transform_te(img) - - -class ImageDataManager(DataManager): - r"""Image data manager. - - Args: - root (str): root path to datasets. - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - k_tfm (int): number of times to apply augmentation to an image - independently. If k_tfm > 1, the transform function will be - applied k_tfm times to an image. This variable will only be - useful for training and is currently valid for image datasets only. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - split_id (int, optional): split id (*0-based*). Default is 0. - combineall (bool, optional): combine train, query and gallery in a dataset for - training. Default is False. - load_train_targets (bool, optional): construct train-loader for target datasets. - Default is False. This is useful for domain adaptation research. - batch_size_train (int, optional): number of images in a training batch. Default is 32. - batch_size_test (int, optional): number of images in a test batch. Default is 32. - workers (int, optional): number of workers. Default is 4. - num_instances (int, optional): number of instances per identity in a batch. - Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - train_sampler (str, optional): sampler. Default is RandomSampler. - train_sampler_t (str, optional): sampler for target train loader. Default is RandomSampler. - cuhk03_labeled (bool, optional): use cuhk03 labeled images. - Default is False (defaul is to use detected images). - cuhk03_classic_split (bool, optional): use the classic split in cuhk03. - Default is False. - market1501_500k (bool, optional): add 500K distractors to the gallery - set in market1501. Default is False. - - Examples:: - - datamanager = torchreid.data.ImageDataManager( - root='path/to/reid-data', - sources='market1501', - height=256, - width=128, - batch_size_train=32, - batch_size_test=100 - ) - - # return train loader of source data - train_loader = datamanager.train_loader - - # return test loader of target data - test_loader = datamanager.test_loader - - # return train loader of target data - train_loader_t = datamanager.train_loader_t - """ - data_type = 'image' - - def __init__( - self, - root='', - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - k_tfm=1, - norm_mean=None, - norm_std=None, - use_gpu=True, - split_id=0, - combineall=False, - load_train_targets=False, - batch_size_train=32, - batch_size_test=32, - workers=4, - num_instances=4, - num_cams=1, - num_datasets=1, - train_sampler='RandomSampler', - train_sampler_t='RandomSampler', - cuhk03_labeled=False, - cuhk03_classic_split=False, - market1501_500k=False, - device_num=-1 - ): - - super(ImageDataManager, self).__init__( - sources=sources, - targets=targets, - height=height, - width=width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std, - use_gpu=use_gpu - ) - - print('=> Loading train (source) dataset') - trainset = [] - for name in self.sources: - trainset_ = init_image_dataset( - name, - transform=self.transform_tr, - k_tfm=k_tfm, - mode='train', - combineall=combineall, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - trainset.append(trainset_) - trainset = sum(trainset) - - self._num_train_pids = trainset.num_train_pids - self._num_train_cams = trainset.num_train_cams - - if device_num == -1 or device_num == 1: - self.train_sampler = build_train_sampler( - trainset.train, - train_sampler, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ) - - else: - self.train_sampler = torch.utils.data.distributed.DistributedSampler(trainset.train) - - self.train_loader = torch.utils.data.DataLoader( - trainset, - sampler=self.train_sampler, - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - # pin_memory=self.use_gpu, - pin_memory=True, - drop_last=True - ) - - self.train_loader_t = None - if load_train_targets: - # check if sources and targets are identical - assert len(set(self.sources) & set(self.targets)) == 0, \ - 'sources={} and targets={} must not have overlap'.format(self.sources, self.targets) - - print('=> Loading train (target) dataset') - trainset_t = [] - for name in self.targets: - trainset_t_ = init_image_dataset( - name, - transform=self.transform_tr, - k_tfm=k_tfm, - mode='train', - combineall=False, # only use the training data - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - trainset_t.append(trainset_t_) - trainset_t = sum(trainset_t) - - self.train_loader_t = torch.utils.data.DataLoader( - trainset_t, - sampler=build_train_sampler( - trainset_t.train, - train_sampler_t, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ), - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=True - ) - - print('=> Loading test (target) dataset') - self.test_loader = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - self.test_dataset = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - - for name in self.targets: - # build query loader - queryset = init_image_dataset( - name, - transform=self.transform_te, - mode='query', - combineall=combineall, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - self.test_loader[name]['query'] = torch.utils.data.DataLoader( - queryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=True, - drop_last=False - ) - - # build gallery loader - galleryset = init_image_dataset( - name, - transform=self.transform_te, - mode='gallery', - combineall=combineall, - verbose=False, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( - galleryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=True, - drop_last=False - ) - - self.test_dataset[name]['query'] = queryset.query - self.test_dataset[name]['gallery'] = galleryset.gallery - - print('\n') - print(' **************** Summary ****************') - print(' source : {}'.format(self.sources)) - print(' # source datasets : {}'.format(len(self.sources))) - print(' # source ids : {}'.format(self.num_train_pids)) - print(' # source images : {}'.format(len(trainset))) - print(' # source cameras : {}'.format(self.num_train_cams)) - if load_train_targets: - print( - ' # target images : {} (unlabeled)'.format(len(trainset_t)) - ) - print(' target : {}'.format(self.targets)) - print(' *****************************************') - print('\n') - - -class VideoDataManager(DataManager): - r"""Video data manager. - - Args: - root (str): root path to datasets. - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - split_id (int, optional): split id (*0-based*). Default is 0. - combineall (bool, optional): combine train, query and gallery in a dataset for - training. Default is False. - batch_size_train (int, optional): number of tracklets in a training batch. Default is 3. - batch_size_test (int, optional): number of tracklets in a test batch. Default is 3. - workers (int, optional): number of workers. Default is 4. - num_instances (int, optional): number of instances per identity in a batch. - Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - train_sampler (str, optional): sampler. Default is RandomSampler. - seq_len (int, optional): how many images to sample in a tracklet. Default is 15. - sample_method (str, optional): how to sample images in a tracklet. Default is "evenly". - Choices are ["evenly", "random", "all"]. "evenly" and "random" will sample ``seq_len`` - images in a tracklet while "all" samples all images in a tracklet, where the batch size - needs to be set to 1. - - Examples:: - - datamanager = torchreid.data.VideoDataManager( - root='path/to/reid-data', - sources='mars', - height=256, - width=128, - batch_size_train=3, - batch_size_test=3, - seq_len=15, - sample_method='evenly' - ) - - # return train loader of source data - train_loader = datamanager.train_loader - - # return test loader of target data - test_loader = datamanager.test_loader - - .. note:: - The current implementation only supports image-like training. Therefore, each image in a - sampled tracklet will undergo independent transformation functions. To achieve tracklet-aware - training, you need to modify the transformation functions for video reid such that each function - applies the same operation to all images in a tracklet to keep consistency. - """ - data_type = 'video' - - def __init__( - self, - root='', - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - norm_mean=None, - norm_std=None, - use_gpu=True, - split_id=0, - combineall=False, - batch_size_train=3, - batch_size_test=3, - workers=4, - num_instances=4, - num_cams=1, - num_datasets=1, - train_sampler='RandomSampler', - seq_len=15, - sample_method='evenly' - ): - - super(VideoDataManager, self).__init__( - sources=sources, - targets=targets, - height=height, - width=width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std, - use_gpu=use_gpu - ) - - print('=> Loading train (source) dataset') - trainset = [] - for name in self.sources: - trainset_ = init_video_dataset( - name, - transform=self.transform_tr, - mode='train', - combineall=combineall, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - trainset.append(trainset_) - trainset = sum(trainset) - - self._num_train_pids = trainset.num_train_pids - self._num_train_cams = trainset.num_train_cams - - train_sampler = build_train_sampler( - trainset.train, - train_sampler, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ) - - self.train_loader = torch.utils.data.DataLoader( - trainset, - sampler=train_sampler, - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=True - ) - - print('=> Loading test (target) dataset') - self.test_loader = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - self.test_dataset = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - - for name in self.targets: - # build query loader - queryset = init_video_dataset( - name, - transform=self.transform_te, - mode='query', - combineall=combineall, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - self.test_loader[name]['query'] = torch.utils.data.DataLoader( - queryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=False - ) - - # build gallery loader - galleryset = init_video_dataset( - name, - transform=self.transform_te, - mode='gallery', - combineall=combineall, - verbose=False, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( - galleryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=False - ) - - self.test_dataset[name]['query'] = queryset.query - self.test_dataset[name]['gallery'] = galleryset.gallery - - print('\n') - print(' **************** Summary ****************') - print(' source : {}'.format(self.sources)) - print(' # source datasets : {}'.format(len(self.sources))) - print(' # source ids : {}'.format(self.num_train_pids)) - print(' # source tracklets : {}'.format(len(trainset))) - print(' # source cameras : {}'.format(self.num_train_cams)) - print(' target : {}'.format(self.targets)) - print(' *****************************************') - print('\n') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py deleted file mode 100644 index 2b18ca7207..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py +++ /dev/null @@ -1,166 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .image import ( - GRID, PRID, CUHK01, CUHK02, CUHK03, MSMT17, CUHKSYSU, VIPeR, SenseReID, - Market1501, DukeMTMCreID, University1652, iLIDS -) -from .video import PRID2011, Mars, DukeMTMCVidReID, iLIDSVID -from .dataset import Dataset, ImageDataset, VideoDataset - -__image_datasets = { - 'market1501': Market1501, - 'cuhk03': CUHK03, - 'dukemtmcreid': DukeMTMCreID, - 'msmt17': MSMT17, - 'viper': VIPeR, - 'grid': GRID, - 'cuhk01': CUHK01, - 'ilids': iLIDS, - 'sensereid': SenseReID, - 'prid': PRID, - 'cuhk02': CUHK02, - 'university1652': University1652, - 'cuhksysu': CUHKSYSU -} - -__video_datasets = { - 'mars': Mars, - 'ilidsvid': iLIDSVID, - 'prid2011': PRID2011, - 'dukemtmcvidreid': DukeMTMCVidReID -} - - -def init_image_dataset(name, **kwargs): - """Initializes an image dataset.""" - avai_datasets = list(__image_datasets.keys()) - if name not in avai_datasets: - raise ValueError( - 'Invalid dataset name. Received "{}", ' - 'but expected to be one of {}'.format(name, avai_datasets) - ) - return __image_datasets[name](**kwargs) - - -def init_video_dataset(name, **kwargs): - """Initializes a video dataset.""" - avai_datasets = list(__video_datasets.keys()) - if name not in avai_datasets: - raise ValueError( - 'Invalid dataset name. Received "{}", ' - 'but expected to be one of {}'.format(name, avai_datasets) - ) - return __video_datasets[name](**kwargs) - - -def register_image_dataset(name, dataset): - """Registers a new image dataset. - - Args: - name (str): key corresponding to the new dataset. - dataset (Dataset): the new dataset class. - - Examples:: - - import torchreid - import NewDataset - torchreid.data.register_image_dataset('new_dataset', NewDataset) - # single dataset case - datamanager = torchreid.data.ImageDataManager( - root='reid-data', - sources='new_dataset' - ) - # multiple dataset case - datamanager = torchreid.data.ImageDataManager( - root='reid-data', - sources=['new_dataset', 'dukemtmcreid'] - ) - """ - global __image_datasets - curr_datasets = list(__image_datasets.keys()) - if name in curr_datasets: - raise ValueError( - 'The given name already exists, please choose ' - 'another name excluding {}'.format(curr_datasets) - ) - __image_datasets[name] = dataset - - -def register_video_dataset(name, dataset): - """Registers a new video dataset. - - Args: - name (str): key corresponding to the new dataset. - dataset (Dataset): the new dataset class. - - Examples:: - - import torchreid - import NewDataset - torchreid.data.register_video_dataset('new_dataset', NewDataset) - # single dataset case - datamanager = torchreid.data.VideoDataManager( - root='reid-data', - sources='new_dataset' - ) - # multiple dataset case - datamanager = torchreid.data.VideoDataManager( - root='reid-data', - sources=['new_dataset', 'ilidsvid'] - ) - """ - global __video_datasets - curr_datasets = list(__video_datasets.keys()) - if name in curr_datasets: - raise ValueError( - 'The given name already exists, please choose ' - 'another name excluding {}'.format(curr_datasets) - ) - __video_datasets[name] = dataset diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py deleted file mode 100644 index f85cb3533e..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py +++ /dev/null @@ -1,529 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import copy -import numpy as np -import os.path as osp -import tarfile -import zipfile -import torch - -from torchreid.utils import read_image, download_url, mkdir_if_missing - - -class Dataset(object): - """An abstract class representing a Dataset. - - This is the base class for ``ImageDataset`` and ``VideoDataset``. - - Args: - train (list): contains tuples of (img_path(s), pid, camid). - query (list): contains tuples of (img_path(s), pid, camid). - gallery (list): contains tuples of (img_path(s), pid, camid). - transform: transform function. - k_tfm (int): number of times to apply augmentation to an image - independently. If k_tfm > 1, the transform function will be - applied k_tfm times to an image. This variable will only be - useful for training and is currently valid for image datasets only. - mode (str): 'train', 'query' or 'gallery'. - combineall (bool): combines train, query and gallery in a - dataset for training. - verbose (bool): show information. - """ - - # junk_pids contains useless person IDs, e.g. background, - # false detections, distractors. These IDs will be ignored - # when combining all images in a dataset for training, i.e. - # combineall=True - _junk_pids = [] - - # Some datasets are only used for training, like CUHK-SYSU - # In this case, "combineall=True" is not used for them - _train_only = False - - def __init__( - self, - train, - query, - gallery, - transform=None, - k_tfm=1, - mode='train', - combineall=False, - verbose=True, - **kwargs - ): - # extend 3-tuple (img_path(s), pid, camid) to - # 4-tuple (img_path(s), pid, camid, dsetid) by - # adding a dataset indicator "dsetid" - if len(train[0]) == 3: - train = [(*items, 0) for items in train] - if len(query[0]) == 3: - query = [(*items, 0) for items in query] - if len(gallery[0]) == 3: - gallery = [(*items, 0) for items in gallery] - - self.train = train - self.query = query - self.gallery = gallery - self.transform = transform - self.k_tfm = k_tfm - self.mode = mode - self.combineall = combineall - self.verbose = verbose - - self.num_train_pids = self.get_num_pids(self.train) - self.num_train_cams = self.get_num_cams(self.train) - self.num_datasets = self.get_num_datasets(self.train) - - if self.combineall: - self.combine_all() - - if self.mode == 'train': - self.data = self.train - elif self.mode == 'query': - self.data = self.query - elif self.mode == 'gallery': - self.data = self.gallery - else: - raise ValueError( - 'Invalid mode. Got {}, but expected to be ' - 'one of [train | query | gallery]'.format(self.mode) - ) - - if self.verbose: - self.show_summary() - - def __getitem__(self, index): - raise NotImplementedError - - def __len__(self): - return len(self.data) - - def __add__(self, other): - """Adds two datasets together (only the train set).""" - train = copy.deepcopy(self.train) - - for img_path, pid, camid, dsetid in other.train: - pid += self.num_train_pids - camid += self.num_train_cams - dsetid += self.num_datasets - train.append((img_path, pid, camid, dsetid)) - - ################################### - # Note that - # 1. set verbose=False to avoid unnecessary print - # 2. set combineall=False because combineall would have been applied - # if it was True for a specific dataset; setting it to True will - # create new IDs that should have already been included - ################################### - if isinstance(train[0][0], str): - return ImageDataset( - train, - self.query, - self.gallery, - transform=self.transform, - mode=self.mode, - combineall=False, - verbose=False - ) - else: - return VideoDataset( - train, - self.query, - self.gallery, - transform=self.transform, - mode=self.mode, - combineall=False, - verbose=False, - seq_len=self.seq_len, - sample_method=self.sample_method - ) - - def __radd__(self, other): - """Supports sum([dataset1, dataset2, dataset3]).""" - if other == 0: - return self - else: - return self.__add__(other) - - def get_num_pids(self, data): - """Returns the number of training person identities. - - Each tuple in data contains (img_path(s), pid, camid, dsetid). - """ - pids = set() - for items in data: - pid = items[1] - pids.add(pid) - return len(pids) - - def get_num_cams(self, data): - """Returns the number of training cameras. - - Each tuple in data contains (img_path(s), pid, camid, dsetid). - """ - cams = set() - for items in data: - camid = items[2] - cams.add(camid) - return len(cams) - - def get_num_datasets(self, data): - """Returns the number of datasets included. - - Each tuple in data contains (img_path(s), pid, camid, dsetid). - """ - dsets = set() - for items in data: - dsetid = items[3] - dsets.add(dsetid) - return len(dsets) - - def show_summary(self): - """Shows dataset statistics.""" - pass - - def combine_all(self): - """Combines train, query and gallery in a dataset for training.""" - if self._train_only: - return - - combined = copy.deepcopy(self.train) - - # relabel pids in gallery (query shares the same scope) - g_pids = set() - for items in self.gallery: - pid = items[1] - if pid in self._junk_pids: - continue - g_pids.add(pid) - pid2label = {pid: i for i, pid in enumerate(g_pids)} - - def _combine_data(data): - for img_path, pid, camid, dsetid in data: - if pid in self._junk_pids: - continue - pid = pid2label[pid] + self.num_train_pids - combined.append((img_path, pid, camid, dsetid)) - - _combine_data(self.query) - _combine_data(self.gallery) - - self.train = combined - self.num_train_pids = self.get_num_pids(self.train) - - def download_dataset(self, dataset_dir, dataset_url): - """Downloads and extracts dataset. - - Args: - dataset_dir (str): dataset directory. - dataset_url (str): url to download dataset. - """ - if osp.exists(dataset_dir): - return - - if dataset_url is None: - raise RuntimeError( - '{} dataset needs to be manually ' - 'prepared, please follow the ' - 'document to prepare this dataset'.format( - self.__class__.__name__ - ) - ) - - print('Creating directory "{}"'.format(dataset_dir)) - mkdir_if_missing(dataset_dir) - fpath = osp.join(dataset_dir, osp.basename(dataset_url)) - - print( - 'Downloading {} dataset to "{}"'.format( - self.__class__.__name__, dataset_dir - ) - ) - download_url(dataset_url, fpath) - - print('Extracting "{}"'.format(fpath)) - try: - tar = tarfile.open(fpath) - tar.extractall(path=dataset_dir) - tar.close() - except: - zip_ref = zipfile.ZipFile(fpath, 'r') - zip_ref.extractall(dataset_dir) - zip_ref.close() - - print('{} dataset is ready'.format(self.__class__.__name__)) - - def check_before_run(self, required_files): - """Checks if required files exist before going deeper. - - Args: - required_files (str or list): string file name(s). - """ - if isinstance(required_files, str): - required_files = [required_files] - - for fpath in required_files: - if not osp.exists(fpath): - raise RuntimeError('"{}" is not found'.format(fpath)) - - def __repr__(self): - num_train_pids = self.get_num_pids(self.train) - num_train_cams = self.get_num_cams(self.train) - - num_query_pids = self.get_num_pids(self.query) - num_query_cams = self.get_num_cams(self.query) - - num_gallery_pids = self.get_num_pids(self.gallery) - num_gallery_cams = self.get_num_cams(self.gallery) - - msg = ' ----------------------------------------\n' \ - ' subset | # ids | # items | # cameras\n' \ - ' ----------------------------------------\n' \ - ' train | {:5d} | {:7d} | {:9d}\n' \ - ' query | {:5d} | {:7d} | {:9d}\n' \ - ' gallery | {:5d} | {:7d} | {:9d}\n' \ - ' ----------------------------------------\n' \ - ' items: images/tracklets for image/video dataset\n'.format( - num_train_pids, len(self.train), num_train_cams, - num_query_pids, len(self.query), num_query_cams, - num_gallery_pids, len(self.gallery), num_gallery_cams - ) - - return msg - - def _transform_image(self, tfm, k_tfm, img0): - """Transforms a raw image (img0) k_tfm times with - the transform function tfm. - """ - img_list = [] - - for k in range(k_tfm): - img_list.append(tfm(img0)) - - img = img_list - if len(img) == 1: - img = img[0] - - return img - - -class ImageDataset(Dataset): - """A base class representing ImageDataset. - - All other image datasets should subclass it. - - ``__getitem__`` returns an image given index. - It will return ``img``, ``pid``, ``camid`` and ``img_path`` - where ``img`` has shape (channel, height, width). As a result, - data in each batch has shape (batch_size, channel, height, width). - """ - - def __init__(self, train, query, gallery, **kwargs): - super(ImageDataset, self).__init__(train, query, gallery, **kwargs) - - def __getitem__(self, index): - img_path, pid, camid, dsetid = self.data[index] - img = read_image(img_path) - if self.transform is not None: - img = self._transform_image(self.transform, self.k_tfm, img) - item = { - 'img': img, - 'pid': pid, - 'camid': camid, - 'impath': img_path, - 'dsetid': dsetid - } - return item - - def show_summary(self): - num_train_pids = self.get_num_pids(self.train) - num_train_cams = self.get_num_cams(self.train) - - num_query_pids = self.get_num_pids(self.query) - num_query_cams = self.get_num_cams(self.query) - - num_gallery_pids = self.get_num_pids(self.gallery) - num_gallery_cams = self.get_num_cams(self.gallery) - - print('=> Loaded {}'.format(self.__class__.__name__)) - print(' ----------------------------------------') - print(' subset | # ids | # images | # cameras') - print(' ----------------------------------------') - print( - ' train | {:5d} | {:8d} | {:9d}'.format( - num_train_pids, len(self.train), num_train_cams - ) - ) - print( - ' query | {:5d} | {:8d} | {:9d}'.format( - num_query_pids, len(self.query), num_query_cams - ) - ) - print( - ' gallery | {:5d} | {:8d} | {:9d}'.format( - num_gallery_pids, len(self.gallery), num_gallery_cams - ) - ) - print(' ----------------------------------------') - - -class VideoDataset(Dataset): - """A base class representing VideoDataset. - - All other video datasets should subclass it. - - ``__getitem__`` returns an image given index. - It will return ``imgs``, ``pid`` and ``camid`` - where ``imgs`` has shape (seq_len, channel, height, width). As a result, - data in each batch has shape (batch_size, seq_len, channel, height, width). - """ - - def __init__( - self, - train, - query, - gallery, - seq_len=15, - sample_method='evenly', - **kwargs - ): - super(VideoDataset, self).__init__(train, query, gallery, **kwargs) - self.seq_len = seq_len - self.sample_method = sample_method - - if self.transform is None: - raise RuntimeError('transform must not be None') - - def __getitem__(self, index): - img_paths, pid, camid, dsetid = self.data[index] - num_imgs = len(img_paths) - - if self.sample_method == 'random': - # Randomly samples seq_len images from a tracklet of length num_imgs, - # if num_imgs is smaller than seq_len, then replicates images - indices = np.arange(num_imgs) - replace = False if num_imgs >= self.seq_len else True - indices = np.random.choice( - indices, size=self.seq_len, replace=replace - ) - # sort indices to keep temporal order (comment it to be order-agnostic) - indices = np.sort(indices) - - elif self.sample_method == 'evenly': - # Evenly samples seq_len images from a tracklet - if num_imgs >= self.seq_len: - num_imgs -= num_imgs % self.seq_len - indices = np.arange(0, num_imgs, num_imgs / self.seq_len) - else: - # if num_imgs is smaller than seq_len, simply replicate the last image - # until the seq_len requirement is satisfied - indices = np.arange(0, num_imgs) - num_pads = self.seq_len - num_imgs - indices = np.concatenate( - [ - indices, - np.ones(num_pads).astype(np.int32) * (num_imgs-1) - ] - ) - assert len(indices) == self.seq_len - - elif self.sample_method == 'all': - # Samples all images in a tracklet. batch_size must be set to 1 - indices = np.arange(num_imgs) - - else: - raise ValueError( - 'Unknown sample method: {}'.format(self.sample_method) - ) - - imgs = [] - for index in indices: - img_path = img_paths[int(index)] - img = read_image(img_path) - if self.transform is not None: - img = self.transform(img) - img = img.unsqueeze(0) # img must be torch.Tensor - imgs.append(img) - imgs = torch.cat(imgs, dim=0) - - item = {'img': imgs, 'pid': pid, 'camid': camid, 'dsetid': dsetid} - - return item - - def show_summary(self): - num_train_pids = self.get_num_pids(self.train) - num_train_cams = self.get_num_cams(self.train) - - num_query_pids = self.get_num_pids(self.query) - num_query_cams = self.get_num_cams(self.query) - - num_gallery_pids = self.get_num_pids(self.gallery) - num_gallery_cams = self.get_num_cams(self.gallery) - - print('=> Loaded {}'.format(self.__class__.__name__)) - print(' -------------------------------------------') - print(' subset | # ids | # tracklets | # cameras') - print(' -------------------------------------------') - print( - ' train | {:5d} | {:11d} | {:9d}'.format( - num_train_pids, len(self.train), num_train_cams - ) - ) - print( - ' query | {:5d} | {:11d} | {:9d}'.format( - num_query_pids, len(self.query), num_query_cams - ) - ) - print( - ' gallery | {:5d} | {:11d} | {:9d}'.format( - num_gallery_pids, len(self.gallery), num_gallery_cams - ) - ) - print(' -------------------------------------------') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py deleted file mode 100644 index a0ed9d5b0e..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py +++ /dev/null @@ -1,62 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .grid import GRID -from .prid import PRID -from .ilids import iLIDS -from .viper import VIPeR -from .cuhk01 import CUHK01 -from .cuhk02 import CUHK02 -from .cuhk03 import CUHK03 -from .msmt17 import MSMT17 -from .cuhksysu import CUHKSYSU -from .sensereid import SenseReID -from .market1501 import Market1501 -from .dukemtmcreid import DukeMTMCreID -from .university1652 import University1652 diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py deleted file mode 100644 index f10f400f8b..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py +++ /dev/null @@ -1,184 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import numpy as np -import os.path as osp -import zipfile - -from torchreid.utils import read_json, write_json - -from ..dataset import ImageDataset - - -class CUHK01(ImageDataset): - """CUHK01. - - Reference: - Li et al. Human Reidentification with Transferred Metric Learning. ACCV 2012. - - URL: ``_ - - Dataset statistics: - - identities: 971. - - images: 3884. - - cameras: 4. - - Note: CUHK01 and CUHK02 overlap. - """ - dataset_dir = 'cuhk01' - dataset_url = None - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.zip_path = osp.join(self.dataset_dir, 'CUHK01.zip') - self.campus_dir = osp.join(self.dataset_dir, 'campus') - self.split_path = osp.join(self.dataset_dir, 'splits.json') - - self.extract_file() - - required_files = [self.dataset_dir, self.campus_dir] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, but expected between 0 and {}' - .format(split_id, - len(splits) - 1) - ) - split = splits[split_id] - - train = split['train'] - query = split['query'] - gallery = split['gallery'] - - train = [tuple(item) for item in train] - query = [tuple(item) for item in query] - gallery = [tuple(item) for item in gallery] - - super(CUHK01, self).__init__(train, query, gallery, **kwargs) - - def extract_file(self): - if not osp.exists(self.campus_dir): - print('Extracting files') - zip_ref = zipfile.ZipFile(self.zip_path, 'r') - zip_ref.extractall(self.dataset_dir) - zip_ref.close() - - def prepare_split(self): - """ - Image name format: 0001001.png, where first four digits represent identity - and last four digits represent cameras. Camera 1&2 are considered the same - view and camera 3&4 are considered the same view. - """ - if not osp.exists(self.split_path): - print('Creating 10 random splits of train ids and test ids') - img_paths = sorted(glob.glob(osp.join(self.campus_dir, '*.png'))) - img_list = [] - pid_container = set() - for img_path in img_paths: - img_name = osp.basename(img_path) - pid = int(img_name[:4]) - 1 - camid = (int(img_name[4:7]) - 1) // 2 # result is either 0 or 1 - img_list.append((img_path, pid, camid)) - pid_container.add(pid) - - num_pids = len(pid_container) - num_train_pids = num_pids // 2 - - splits = [] - for _ in range(10): - order = np.arange(num_pids) - np.random.shuffle(order) - train_idxs = order[:num_train_pids] - train_idxs = np.sort(train_idxs) - idx2label = { - idx: label - for label, idx in enumerate(train_idxs) - } - - train, test_a, test_b = [], [], [] - for img_path, pid, camid in img_list: - if pid in train_idxs: - train.append((img_path, idx2label[pid], camid)) - else: - if camid == 0: - test_a.append((img_path, pid, camid)) - else: - test_b.append((img_path, pid, camid)) - - # use cameraA as query and cameraB as gallery - split = { - 'train': train, - 'query': test_a, - 'gallery': test_b, - 'num_train_pids': num_train_pids, - 'num_query_pids': num_pids - num_train_pids, - 'num_gallery_pids': num_pids - num_train_pids - } - splits.append(split) - - # use cameraB as query and cameraA as gallery - split = { - 'train': train, - 'query': test_b, - 'gallery': test_a, - 'num_train_pids': num_train_pids, - 'num_query_pids': num_pids - num_train_pids, - 'num_gallery_pids': num_pids - num_train_pids - } - splits.append(split) - - print('Totally {} splits are created'.format(len(splits))) - write_json(splits, self.split_path) - print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py deleted file mode 100644 index 7eee2aa965..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py +++ /dev/null @@ -1,144 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import os.path as osp - -from ..dataset import ImageDataset - - -class CUHK02(ImageDataset): - """CUHK02. - - Reference: - Li and Wang. Locally Aligned Feature Transforms across Views. CVPR 2013. - - URL: ``_ - - Dataset statistics: - - 5 camera view pairs each with two cameras - - 971, 306, 107, 193 and 239 identities from P1 - P5 - - totally 1,816 identities - - image format is png - - Protocol: Use P1 - P4 for training and P5 for evaluation. - - Note: CUHK01 and CUHK02 overlap. - """ - dataset_dir = 'cuhk02' - cam_pairs = ['P1', 'P2', 'P3', 'P4', 'P5'] - test_cam_pair = 'P5' - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir, 'Dataset') - - required_files = [self.dataset_dir] - self.check_before_run(required_files) - - train, query, gallery = self.get_data_list() - - super(CUHK02, self).__init__(train, query, gallery, **kwargs) - - def get_data_list(self): - num_train_pids, camid = 0, 0 - train, query, gallery = [], [], [] - - for cam_pair in self.cam_pairs: - cam_pair_dir = osp.join(self.dataset_dir, cam_pair) - - cam1_dir = osp.join(cam_pair_dir, 'cam1') - cam2_dir = osp.join(cam_pair_dir, 'cam2') - - impaths1 = glob.glob(osp.join(cam1_dir, '*.png')) - impaths2 = glob.glob(osp.join(cam2_dir, '*.png')) - - if cam_pair == self.test_cam_pair: - # add images to query - for impath in impaths1: - pid = osp.basename(impath).split('_')[0] - pid = int(pid) - query.append((impath, pid, camid)) - camid += 1 - - # add images to gallery - for impath in impaths2: - pid = osp.basename(impath).split('_')[0] - pid = int(pid) - gallery.append((impath, pid, camid)) - camid += 1 - - else: - pids1 = [ - osp.basename(impath).split('_')[0] for impath in impaths1 - ] - pids2 = [ - osp.basename(impath).split('_')[0] for impath in impaths2 - ] - pids = set(pids1 + pids2) - pid2label = { - pid: label + num_train_pids - for label, pid in enumerate(pids) - } - - # add images to train from cam1 - for impath in impaths1: - pid = osp.basename(impath).split('_')[0] - pid = pid2label[pid] - train.append((impath, pid, camid)) - camid += 1 - - # add images to train from cam2 - for impath in impaths2: - pid = osp.basename(impath).split('_')[0] - pid = pid2label[pid] - train.append((impath, pid, camid)) - camid += 1 - num_train_pids += len(pids) - - return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py deleted file mode 100644 index 8917929bdc..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py +++ /dev/null @@ -1,354 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import os.path as osp - -from torchreid.utils import read_json, write_json, mkdir_if_missing - -from ..dataset import ImageDataset - - -class CUHK03(ImageDataset): - """CUHK03. - - Reference: - Li et al. DeepReID: Deep Filter Pairing Neural Network for Person Re-identification. CVPR 2014. - - URL: ``_ - - Dataset statistics: - - identities: 1360. - - images: 13164. - - cameras: 6. - - splits: 20 (classic). - """ - dataset_dir = 'cuhk03' - dataset_url = None - - def __init__( - self, - root='', - split_id=0, - cuhk03_labeled=False, - cuhk03_classic_split=False, - **kwargs - ): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.data_dir = osp.join(self.dataset_dir, 'cuhk03_release') - self.raw_mat_path = osp.join(self.data_dir, 'cuhk-03.mat') - - self.imgs_detected_dir = osp.join(self.dataset_dir, 'images_detected') - self.imgs_labeled_dir = osp.join(self.dataset_dir, 'images_labeled') - - self.split_classic_det_json_path = osp.join( - self.dataset_dir, 'splits_classic_detected.json' - ) - self.split_classic_lab_json_path = osp.join( - self.dataset_dir, 'splits_classic_labeled.json' - ) - - self.split_new_det_json_path = osp.join( - self.dataset_dir, 'splits_new_detected.json' - ) - self.split_new_lab_json_path = osp.join( - self.dataset_dir, 'splits_new_labeled.json' - ) - - self.split_new_det_mat_path = osp.join( - self.dataset_dir, 'cuhk03_new_protocol_config_detected.mat' - ) - self.split_new_lab_mat_path = osp.join( - self.dataset_dir, 'cuhk03_new_protocol_config_labeled.mat' - ) - - required_files = [ - self.dataset_dir, self.data_dir, self.raw_mat_path, - self.split_new_det_mat_path, self.split_new_lab_mat_path - ] - self.check_before_run(required_files) - - self.preprocess_split() - - if cuhk03_labeled: - split_path = self.split_classic_lab_json_path if cuhk03_classic_split else self.split_new_lab_json_path - else: - split_path = self.split_classic_det_json_path if cuhk03_classic_split else self.split_new_det_json_path - - splits = read_json(split_path) - assert split_id < len( - splits - ), 'Condition split_id ({}) < len(splits) ({}) is false'.format( - split_id, len(splits) - ) - split = splits[split_id] - - train = split['train'] - query = split['query'] - gallery = split['gallery'] - - super(CUHK03, self).__init__(train, query, gallery, **kwargs) - - def preprocess_split(self): - # This function is a bit complex and ugly, what it does is - # 1. extract data from cuhk-03.mat and save as png images - # 2. create 20 classic splits (Li et al. CVPR'14) - # 3. create new split (Zhong et al. CVPR'17) - if osp.exists(self.imgs_labeled_dir) \ - and osp.exists(self.imgs_detected_dir) \ - and osp.exists(self.split_classic_det_json_path) \ - and osp.exists(self.split_classic_lab_json_path) \ - and osp.exists(self.split_new_det_json_path) \ - and osp.exists(self.split_new_lab_json_path): - return - - import h5py - import imageio - from scipy.io import loadmat - - mkdir_if_missing(self.imgs_detected_dir) - mkdir_if_missing(self.imgs_labeled_dir) - - print( - 'Extract image data from "{}" and save as png'.format( - self.raw_mat_path - ) - ) - mat = h5py.File(self.raw_mat_path, 'r') - - def _deref(ref): - return mat[ref][:].T - - def _process_images(img_refs, campid, pid, save_dir): - img_paths = [] # Note: some persons only have images for one view - for imgid, img_ref in enumerate(img_refs): - img = _deref(img_ref) - if img.size == 0 or img.ndim < 3: - continue # skip empty cell - # images are saved with the following format, index-1 (ensure uniqueness) - # campid: index of camera pair (1-5) - # pid: index of person in 'campid'-th camera pair - # viewid: index of view, {1, 2} - # imgid: index of image, (1-10) - viewid = 1 if imgid < 5 else 2 - img_name = '{:01d}_{:03d}_{:01d}_{:02d}.png'.format( - campid + 1, pid + 1, viewid, imgid + 1 - ) - img_path = osp.join(save_dir, img_name) - if not osp.isfile(img_path): - imageio.imwrite(img_path, img) - img_paths.append(img_path) - return img_paths - - def _extract_img(image_type): - print('Processing {} images ...'.format(image_type)) - meta_data = [] - imgs_dir = self.imgs_detected_dir if image_type == 'detected' else self.imgs_labeled_dir - for campid, camp_ref in enumerate(mat[image_type][0]): - camp = _deref(camp_ref) - num_pids = camp.shape[0] - for pid in range(num_pids): - img_paths = _process_images( - camp[pid, :], campid, pid, imgs_dir - ) - assert len(img_paths) > 0, \ - 'campid{}-pid{} has no images'.format(campid, pid) - meta_data.append((campid + 1, pid + 1, img_paths)) - print( - '- done camera pair {} with {} identities'.format( - campid + 1, num_pids - ) - ) - return meta_data - - meta_detected = _extract_img('detected') - meta_labeled = _extract_img('labeled') - - def _extract_classic_split(meta_data, test_split): - train, test = [], [] - num_train_pids, num_test_pids = 0, 0 - num_train_imgs, num_test_imgs = 0, 0 - for i, (campid, pid, img_paths) in enumerate(meta_data): - - if [campid, pid] in test_split: - for img_path in img_paths: - camid = int( - osp.basename(img_path).split('_')[2] - ) - 1 # make it 0-based - test.append((img_path, num_test_pids, camid)) - num_test_pids += 1 - num_test_imgs += len(img_paths) - else: - for img_path in img_paths: - camid = int( - osp.basename(img_path).split('_')[2] - ) - 1 # make it 0-based - train.append((img_path, num_train_pids, camid)) - num_train_pids += 1 - num_train_imgs += len(img_paths) - return train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs - - print('Creating classic splits (# = 20) ...') - splits_classic_det, splits_classic_lab = [], [] - for split_ref in mat['testsets'][0]: - test_split = _deref(split_ref).tolist() - - # create split for detected images - train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs = \ - _extract_classic_split(meta_detected, test_split) - splits_classic_det.append( - { - 'train': train, - 'query': test, - 'gallery': test, - 'num_train_pids': num_train_pids, - 'num_train_imgs': num_train_imgs, - 'num_query_pids': num_test_pids, - 'num_query_imgs': num_test_imgs, - 'num_gallery_pids': num_test_pids, - 'num_gallery_imgs': num_test_imgs - } - ) - - # create split for labeled images - train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs = \ - _extract_classic_split(meta_labeled, test_split) - splits_classic_lab.append( - { - 'train': train, - 'query': test, - 'gallery': test, - 'num_train_pids': num_train_pids, - 'num_train_imgs': num_train_imgs, - 'num_query_pids': num_test_pids, - 'num_query_imgs': num_test_imgs, - 'num_gallery_pids': num_test_pids, - 'num_gallery_imgs': num_test_imgs - } - ) - - write_json(splits_classic_det, self.split_classic_det_json_path) - write_json(splits_classic_lab, self.split_classic_lab_json_path) - - def _extract_set(filelist, pids, pid2label, idxs, img_dir, relabel): - tmp_set = [] - unique_pids = set() - for idx in idxs: - img_name = filelist[idx][0] - camid = int(img_name.split('_')[2]) - 1 # make it 0-based - pid = pids[idx] - if relabel: - pid = pid2label[pid] - img_path = osp.join(img_dir, img_name) - tmp_set.append((img_path, int(pid), camid)) - unique_pids.add(pid) - return tmp_set, len(unique_pids), len(idxs) - - def _extract_new_split(split_dict, img_dir): - train_idxs = split_dict['train_idx'].flatten() - 1 # index-0 - pids = split_dict['labels'].flatten() - train_pids = set(pids[train_idxs]) - pid2label = {pid: label for label, pid in enumerate(train_pids)} - query_idxs = split_dict['query_idx'].flatten() - 1 - gallery_idxs = split_dict['gallery_idx'].flatten() - 1 - filelist = split_dict['filelist'].flatten() - train_info = _extract_set( - filelist, pids, pid2label, train_idxs, img_dir, relabel=True - ) - query_info = _extract_set( - filelist, pids, pid2label, query_idxs, img_dir, relabel=False - ) - gallery_info = _extract_set( - filelist, - pids, - pid2label, - gallery_idxs, - img_dir, - relabel=False - ) - return train_info, query_info, gallery_info - - print('Creating new split for detected images (767/700) ...') - train_info, query_info, gallery_info = _extract_new_split( - loadmat(self.split_new_det_mat_path), self.imgs_detected_dir - ) - split = [ - { - 'train': train_info[0], - 'query': query_info[0], - 'gallery': gallery_info[0], - 'num_train_pids': train_info[1], - 'num_train_imgs': train_info[2], - 'num_query_pids': query_info[1], - 'num_query_imgs': query_info[2], - 'num_gallery_pids': gallery_info[1], - 'num_gallery_imgs': gallery_info[2] - } - ] - write_json(split, self.split_new_det_json_path) - - print('Creating new split for labeled images (767/700) ...') - train_info, query_info, gallery_info = _extract_new_split( - loadmat(self.split_new_lab_mat_path), self.imgs_labeled_dir - ) - split = [ - { - 'train': train_info[0], - 'query': query_info[0], - 'gallery': gallery_info[0], - 'num_train_pids': train_info[1], - 'num_train_imgs': train_info[2], - 'num_query_pids': query_info[1], - 'num_query_imgs': query_info[2], - 'num_gallery_pids': gallery_info[1], - 'num_gallery_imgs': gallery_info[2] - } - ] - write_json(split, self.split_new_lab_json_path) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py deleted file mode 100644 index 1e8e25977a..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py +++ /dev/null @@ -1,107 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import copy -import glob -import os.path as osp - -from ..dataset import ImageDataset - - -class CUHKSYSU(ImageDataset): - """CUHKSYSU. - - This dataset can only be used for model training. - - Reference: - Xiao et al. End-to-end deep learning for person search. - - URL: ``_ - - Dataset statistics: - - identities: 11,934 - - images: 34,574 - """ - _train_only = True - dataset_dir = 'cuhksysu' - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.data_dir = osp.join(self.dataset_dir, 'cropped_images') - - # image name format: p11422_s16929_1.jpg - train = self.process_dir(self.data_dir) - query = [copy.deepcopy(train[0])] - gallery = [copy.deepcopy(train[0])] - - super(CUHKSYSU, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dirname): - img_paths = glob.glob(osp.join(dirname, '*.jpg')) - # num_imgs = len(img_paths) - - # get all identities: - pid_container = set() - for img_path in img_paths: - img_name = osp.basename(img_path) - pid = img_name.split('_')[0] - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - - # num_pids = len(pid_container) - - # extract data - data = [] - for img_path in img_paths: - img_name = osp.basename(img_path) - pid = img_name.split('_')[0] - label = pid2label[pid] - data.append((img_path, label, 0)) # dummy camera id - - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py deleted file mode 100644 index 62ca860d6b..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py +++ /dev/null @@ -1,115 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import re -import glob -import os.path as osp - -from ..dataset import ImageDataset - - -class DukeMTMCreID(ImageDataset): - """DukeMTMC-reID. - - Reference: - - Ristani et al. Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. ECCVW 2016. - - Zheng et al. Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro. ICCV 2017. - - URL: ``_ - - Dataset statistics: - - identities: 1404 (train + query). - - images:16522 (train) + 2228 (query) + 17661 (gallery). - - cameras: 8. - """ - dataset_dir = 'dukemtmc-reid' - dataset_url = 'http://vision.cs.duke.edu/DukeMTMC/data/misc/DukeMTMC-reID.zip' - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - self.train_dir = osp.join( - self.dataset_dir, 'DukeMTMC-reID/bounding_box_train' - ) - self.query_dir = osp.join(self.dataset_dir, 'DukeMTMC-reID/query') - self.gallery_dir = osp.join( - self.dataset_dir, 'DukeMTMC-reID/bounding_box_test' - ) - - required_files = [ - self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir - ] - self.check_before_run(required_files) - - train = self.process_dir(self.train_dir, relabel=True) - query = self.process_dir(self.query_dir, relabel=False) - gallery = self.process_dir(self.gallery_dir, relabel=False) - - super(DukeMTMCreID, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path, relabel=False): - img_paths = glob.glob(osp.join(dir_path, '*.jpg')) - pattern = re.compile(r'([-\d]+)_c(\d)') - - pid_container = set() - for img_path in img_paths: - pid, _ = map(int, pattern.search(img_path).groups()) - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - - data = [] - for img_path in img_paths: - pid, camid = map(int, pattern.search(img_path).groups()) - assert 1 <= camid <= 8 - camid -= 1 # index starts from 0 - if relabel: - pid = pid2label[pid] - data.append((img_path, pid, camid)) - - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py deleted file mode 100644 index bfe897b649..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py +++ /dev/null @@ -1,178 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import os.path as osp -from scipy.io import loadmat - -from torchreid.utils import read_json, write_json - -from ..dataset import ImageDataset - - -class GRID(ImageDataset): - """GRID. - - Reference: - Loy et al. Multi-camera activity correlation analysis. CVPR 2009. - - URL: ``_ - - Dataset statistics: - - identities: 250. - - images: 1275. - - cameras: 8. - """ - dataset_dir = 'grid' - dataset_url = 'http://personal.ie.cuhk.edu.hk/~ccloy/files/datasets/underground_reid.zip' - _junk_pids = [0] - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.probe_path = osp.join( - self.dataset_dir, 'underground_reid', 'probe' - ) - self.gallery_path = osp.join( - self.dataset_dir, 'underground_reid', 'gallery' - ) - self.split_mat_path = osp.join( - self.dataset_dir, 'underground_reid', 'features_and_partitions.mat' - ) - self.split_path = osp.join(self.dataset_dir, 'splits.json') - - required_files = [ - self.dataset_dir, self.probe_path, self.gallery_path, - self.split_mat_path - ] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, ' - 'but expected between 0 and {}'.format( - split_id, - len(splits) - 1 - ) - ) - split = splits[split_id] - - train = split['train'] - query = split['query'] - gallery = split['gallery'] - - train = [tuple(item) for item in train] - query = [tuple(item) for item in query] - gallery = [tuple(item) for item in gallery] - - super(GRID, self).__init__(train, query, gallery, **kwargs) - - def prepare_split(self): - if not osp.exists(self.split_path): - print('Creating 10 random splits') - split_mat = loadmat(self.split_mat_path) - trainIdxAll = split_mat['trainIdxAll'][0] # length = 10 - probe_img_paths = sorted( - glob.glob(osp.join(self.probe_path, '*.jpeg')) - ) - gallery_img_paths = sorted( - glob.glob(osp.join(self.gallery_path, '*.jpeg')) - ) - - splits = [] - for split_idx in range(10): - train_idxs = trainIdxAll[split_idx][0][0][2][0].tolist() - assert len(train_idxs) == 125 - idx2label = { - idx: label - for label, idx in enumerate(train_idxs) - } - - train, query, gallery = [], [], [] - - # processing probe folder - for img_path in probe_img_paths: - img_name = osp.basename(img_path) - img_idx = int(img_name.split('_')[0]) - camid = int( - img_name.split('_')[1] - ) - 1 # index starts from 0 - if img_idx in train_idxs: - train.append((img_path, idx2label[img_idx], camid)) - else: - query.append((img_path, img_idx, camid)) - - # process gallery folder - for img_path in gallery_img_paths: - img_name = osp.basename(img_path) - img_idx = int(img_name.split('_')[0]) - camid = int( - img_name.split('_')[1] - ) - 1 # index starts from 0 - if img_idx in train_idxs: - train.append((img_path, idx2label[img_idx], camid)) - else: - gallery.append((img_path, img_idx, camid)) - - split = { - 'train': train, - 'query': query, - 'gallery': gallery, - 'num_train_pids': 125, - 'num_query_pids': 125, - 'num_gallery_pids': 900 - } - splits.append(split) - - print('Totally {} splits are created'.format(len(splits))) - write_json(splits, self.split_path) - print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py deleted file mode 100644 index e2a97265c3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py +++ /dev/null @@ -1,182 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import copy -import glob -import random -import os.path as osp -from collections import defaultdict - -from torchreid.utils import read_json, write_json - -from ..dataset import ImageDataset - - -class iLIDS(ImageDataset): - """QMUL-iLIDS. - - Reference: - Zheng et al. Associating Groups of People. BMVC 2009. - - Dataset statistics: - - identities: 119. - - images: 476. - - cameras: 8 (not explicitly provided). - """ - dataset_dir = 'ilids' - dataset_url = 'http://www.eecs.qmul.ac.uk/~jason/data/i-LIDS_Pedestrian.tgz' - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.data_dir = osp.join(self.dataset_dir, 'i-LIDS_Pedestrian/Persons') - self.split_path = osp.join(self.dataset_dir, 'splits.json') - - required_files = [self.dataset_dir, self.data_dir] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, but ' - 'expected between 0 and {}'.format(split_id, - len(splits) - 1) - ) - split = splits[split_id] - - train, query, gallery = self.process_split(split) - - super(iLIDS, self).__init__(train, query, gallery, **kwargs) - - def prepare_split(self): - if not osp.exists(self.split_path): - print('Creating splits ...') - - paths = glob.glob(osp.join(self.data_dir, '*.jpg')) - img_names = [osp.basename(path) for path in paths] - num_imgs = len(img_names) - assert num_imgs == 476, 'There should be 476 images, but ' \ - 'got {}, please check the data'.format(num_imgs) - - # store image names - # image naming format: - # the first four digits denote the person ID - # the last four digits denote the sequence index - pid_dict = defaultdict(list) - for img_name in img_names: - pid = int(img_name[:4]) - pid_dict[pid].append(img_name) - pids = list(pid_dict.keys()) - num_pids = len(pids) - assert num_pids == 119, 'There should be 119 identities, ' \ - 'but got {}, please check the data'.format(num_pids) - - num_train_pids = int(num_pids * 0.5) - - splits = [] - for _ in range(10): - # randomly choose num_train_pids train IDs and the rest for test IDs - pids_copy = copy.deepcopy(pids) - random.shuffle(pids_copy) - train_pids = pids_copy[:num_train_pids] - test_pids = pids_copy[num_train_pids:] - - train = [] - query = [] - gallery = [] - - # for train IDs, all images are used in the train set. - for pid in train_pids: - img_names = pid_dict[pid] - train.extend(img_names) - - # for each test ID, randomly choose two images, one for - # query and the other one for gallery. - for pid in test_pids: - img_names = pid_dict[pid] - samples = random.sample(img_names, 2) - query.append(samples[0]) - gallery.append(samples[1]) - - split = {'train': train, 'query': query, 'gallery': gallery} - splits.append(split) - - print('Totally {} splits are created'.format(len(splits))) - write_json(splits, self.split_path) - print('Split file is saved to {}'.format(self.split_path)) - - def get_pid2label(self, img_names): - pid_container = set() - for img_name in img_names: - pid = int(img_name[:4]) - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - return pid2label - - def parse_img_names(self, img_names, pid2label=None): - data = [] - - for img_name in img_names: - pid = int(img_name[:4]) - if pid2label is not None: - pid = pid2label[pid] - camid = int(img_name[4:7]) - 1 # 0-based - img_path = osp.join(self.data_dir, img_name) - data.append((img_path, pid, camid)) - - return data - - def process_split(self, split): - train_pid2label = self.get_pid2label(split['train']) - train = self.parse_img_names(split['train'], train_pid2label) - query = self.parse_img_names(split['query']) - gallery = self.parse_img_names(split['gallery']) - return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py deleted file mode 100644 index 7cdf598f6e..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py +++ /dev/null @@ -1,133 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import re -import glob -import os.path as osp -import warnings - -from ..dataset import ImageDataset - - -class Market1501(ImageDataset): - """Market1501. - - Reference: - Zheng et al. Scalable Person Re-identification: A Benchmark. ICCV 2015. - - URL: ``_ - - Dataset statistics: - - identities: 1501 (+1 for background). - - images: 12936 (train) + 3368 (query) + 15913 (gallery). - """ - _junk_pids = [0, -1] - dataset_dir = 'market1501' - - def __init__(self, root='', market1501_500k=False, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - - # allow alternative directory structure - self.data_dir = self.dataset_dir - data_dir = osp.join(self.data_dir, 'Market-1501-v15.09.15') - if osp.isdir(data_dir): - self.data_dir = data_dir - else: - warnings.warn( - 'The current data structure is deprecated. Please ' - 'put data folders such as "bounding_box_train" under ' - '"Market-1501-v15.09.15".' - ) - - self.train_dir = osp.join(self.data_dir, 'bounding_box_train') - self.query_dir = osp.join(self.data_dir, 'query') - self.gallery_dir = osp.join(self.data_dir, 'bounding_box_test') - self.extra_gallery_dir = osp.join(self.data_dir, 'images') - self.market1501_500k = market1501_500k - - required_files = [ - self.data_dir, self.train_dir, self.query_dir, self.gallery_dir - ] - if self.market1501_500k: - required_files.append(self.extra_gallery_dir) - self.check_before_run(required_files) - - train = self.process_dir(self.train_dir, relabel=True) - query = self.process_dir(self.query_dir, relabel=False) - gallery = self.process_dir(self.gallery_dir, relabel=False) - if self.market1501_500k: - gallery += self.process_dir(self.extra_gallery_dir, relabel=False) - - super(Market1501, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path, relabel=False): - img_paths = glob.glob(osp.join(dir_path, '*.jpg')) - pattern = re.compile(r'([-\d]+)_c(\d)') - - pid_container = set() - for img_path in img_paths: - pid, _ = map(int, pattern.search(img_path).groups()) - if pid == -1: - continue # junk images are just ignored - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - - data = [] - for img_path in img_paths: - pid, camid = map(int, pattern.search(img_path).groups()) - if pid == -1: - continue # junk images are just ignored - assert 0 <= pid <= 1501 # pid == 0 means background - assert 1 <= camid <= 6 - camid -= 1 # index starts from 0 - if relabel: - pid = pid2label[pid] - data.append((img_path, pid, camid)) - - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py deleted file mode 100644 index 6f5559bea3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py +++ /dev/null @@ -1,145 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import os.path as osp - -from ..dataset import ImageDataset - -# Log -# 22.01.2019 -# - add v2 -# - v1 and v2 differ in dir names -# - note that faces in v2 are blurred -TRAIN_DIR_KEY = 'train_dir' -TEST_DIR_KEY = 'test_dir' -VERSION_DICT = { - 'MSMT17_V1': { - TRAIN_DIR_KEY: 'train', - TEST_DIR_KEY: 'test', - }, - 'MSMT17_V2': { - TRAIN_DIR_KEY: 'mask_train_v2', - TEST_DIR_KEY: 'mask_test_v2', - } -} - - -class MSMT17(ImageDataset): - """MSMT17. - - Reference: - Wei et al. Person Transfer GAN to Bridge Domain Gap for Person Re-Identification. CVPR 2018. - - URL: ``_ - - Dataset statistics: - - identities: 4101. - - images: 32621 (train) + 11659 (query) + 82161 (gallery). - - cameras: 15. - """ - dataset_dir = 'msmt17' - dataset_url = None - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - has_main_dir = False - for main_dir in VERSION_DICT: - if osp.exists(osp.join(self.dataset_dir, main_dir)): - train_dir = VERSION_DICT[main_dir][TRAIN_DIR_KEY] - test_dir = VERSION_DICT[main_dir][TEST_DIR_KEY] - has_main_dir = True - break - assert has_main_dir, 'Dataset folder not found' - - self.train_dir = osp.join(self.dataset_dir, main_dir, train_dir) - self.test_dir = osp.join(self.dataset_dir, main_dir, test_dir) - self.list_train_path = osp.join( - self.dataset_dir, main_dir, 'list_train.txt' - ) - self.list_val_path = osp.join( - self.dataset_dir, main_dir, 'list_val.txt' - ) - self.list_query_path = osp.join( - self.dataset_dir, main_dir, 'list_query.txt' - ) - self.list_gallery_path = osp.join( - self.dataset_dir, main_dir, 'list_gallery.txt' - ) - - required_files = [self.dataset_dir, self.train_dir, self.test_dir] - self.check_before_run(required_files) - - train = self.process_dir(self.train_dir, self.list_train_path) - val = self.process_dir(self.train_dir, self.list_val_path) - query = self.process_dir(self.test_dir, self.list_query_path) - gallery = self.process_dir(self.test_dir, self.list_gallery_path) - - # Note: to fairly compare with published methods on the conventional ReID setting, - # do not add val images to the training set. - if 'combineall' in kwargs and kwargs['combineall']: - train += val - - super(MSMT17, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path, list_path): - with open(list_path, 'r') as txt: - lines = txt.readlines() - - data = [] - - for img_idx, img_info in enumerate(lines): - img_path, pid = img_info.split(' ') - pid = int(pid) # no need to relabel - camid = int(img_path.split('_')[2]) - 1 # index starts from 0 - img_path = osp.join(dir_path, img_path) - data.append((img_path, pid, camid)) - - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py deleted file mode 100644 index cebb77d1b9..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py +++ /dev/null @@ -1,154 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import random -import os.path as osp - -from torchreid.utils import read_json, write_json - -from ..dataset import ImageDataset - - -class PRID(ImageDataset): - """PRID (single-shot version of prid-2011) - - Reference: - Hirzer et al. Person Re-Identification by Descriptive and Discriminative - Classification. SCIA 2011. - - URL: ``_ - - Dataset statistics: - - Two views. - - View A captures 385 identities. - - View B captures 749 identities. - - 200 identities appear in both views (index starts from 1 to 200). - """ - dataset_dir = 'prid2011' - dataset_url = None - _junk_pids = list(range(201, 750)) - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.cam_a_dir = osp.join( - self.dataset_dir, 'prid_2011', 'single_shot', 'cam_a' - ) - self.cam_b_dir = osp.join( - self.dataset_dir, 'prid_2011', 'single_shot', 'cam_b' - ) - self.split_path = osp.join(self.dataset_dir, 'splits_single_shot.json') - - required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, but expected between 0 and {}' - .format(split_id, - len(splits) - 1) - ) - split = splits[split_id] - - train, query, gallery = self.process_split(split) - - super(PRID, self).__init__(train, query, gallery, **kwargs) - - def prepare_split(self): - if not osp.exists(self.split_path): - print('Creating splits ...') - - splits = [] - for _ in range(10): - # randomly sample 100 IDs for train and use the rest 100 IDs for test - # (note: there are only 200 IDs appearing in both views) - pids = [i for i in range(1, 201)] - train_pids = random.sample(pids, 100) - train_pids.sort() - test_pids = [i for i in pids if i not in train_pids] - split = {'train': train_pids, 'test': test_pids} - splits.append(split) - - print('Totally {} splits are created'.format(len(splits))) - write_json(splits, self.split_path) - print('Split file is saved to {}'.format(self.split_path)) - - def process_split(self, split): - train_pids = split['train'] - test_pids = split['test'] - - train_pid2label = {pid: label for label, pid in enumerate(train_pids)} - - # train - train = [] - for pid in train_pids: - img_name = 'person_' + str(pid).zfill(4) + '.png' - pid = train_pid2label[pid] - img_a_path = osp.join(self.cam_a_dir, img_name) - train.append((img_a_path, pid, 0)) - img_b_path = osp.join(self.cam_b_dir, img_name) - train.append((img_b_path, pid, 1)) - - # query and gallery - query, gallery = [], [] - for pid in test_pids: - img_name = 'person_' + str(pid).zfill(4) + '.png' - img_a_path = osp.join(self.cam_a_dir, img_name) - query.append((img_a_path, pid, 0)) - img_b_path = osp.join(self.cam_b_dir, img_name) - gallery.append((img_b_path, pid, 1)) - for pid in range(201, 750): - img_name = 'person_' + str(pid).zfill(4) + '.png' - img_b_path = osp.join(self.cam_b_dir, img_name) - gallery.append((img_b_path, pid, 1)) - - return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py deleted file mode 100644 index 22547885c6..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py +++ /dev/null @@ -1,117 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import copy -import glob -import os.path as osp - -from ..dataset import ImageDataset - - -class SenseReID(ImageDataset): - """SenseReID. - - This dataset is used for test purpose only. - - Reference: - Zhao et al. Spindle Net: Person Re-identification with Human Body - Region Guided Feature Decomposition and Fusion. CVPR 2017. - - URL: ``_ - - Dataset statistics: - - query: 522 ids, 1040 images. - - gallery: 1717 ids, 3388 images. - """ - dataset_dir = 'sensereid' - dataset_url = None - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.query_dir = osp.join(self.dataset_dir, 'SenseReID', 'test_probe') - self.gallery_dir = osp.join( - self.dataset_dir, 'SenseReID', 'test_gallery' - ) - - required_files = [self.dataset_dir, self.query_dir, self.gallery_dir] - self.check_before_run(required_files) - - query = self.process_dir(self.query_dir) - gallery = self.process_dir(self.gallery_dir) - - # relabel - g_pids = set() - for _, pid, _ in gallery: - g_pids.add(pid) - pid2label = {pid: i for i, pid in enumerate(g_pids)} - - query = [ - (img_path, pid2label[pid], camid) for img_path, pid, camid in query - ] - gallery = [ - (img_path, pid2label[pid], camid) - for img_path, pid, camid in gallery - ] - train = copy.deepcopy(query) + copy.deepcopy(gallery) # dummy variable - - super(SenseReID, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path): - img_paths = glob.glob(osp.join(dir_path, '*.jpg')) - data = [] - - for img_path in img_paths: - img_name = osp.splitext(osp.basename(img_path))[0] - pid, camid = img_name.split('_') - pid, camid = int(pid), int(camid) - data.append((img_path, pid, camid)) - - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py deleted file mode 100644 index bd950fe3d8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py +++ /dev/null @@ -1,157 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import os -import glob -import os.path as osp -import gdown - -from ..dataset import ImageDataset - - -class University1652(ImageDataset): - """University-1652. - - Reference: - - Zheng et al. University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization. ACM MM 2020. - - URL: ``_ - OneDrive: - https://studentutsedu-my.sharepoint.com/:u:/g/personal/12639605_student_uts_edu_au/Ecrz6xK-PcdCjFdpNb0T0s8B_9J5ynaUy3q63_XumjJyrA?e=z4hpcz - [Backup] GoogleDrive: - https://drive.google.com/file/d/1iVnP4gjw-iHXa0KerZQ1IfIO0i1jADsR/view?usp=sharing - [Backup] Baidu Yun: - https://pan.baidu.com/s/1H_wBnWwikKbaBY1pMPjoqQ password: hrqp - - Dataset statistics: - - buildings: 1652 (train + query). - - The dataset split is as follows: - | Split | #imgs | #buildings | #universities| - | -------- | ----- | ----| ----| - | Training | 50,218 | 701 | 33 | - | Query_drone | 37,855 | 701 | 39 | - | Query_satellite | 701 | 701 | 39| - | Query_ground | 2,579 | 701 | 39| - | Gallery_drone | 51,355 | 951 | 39| - | Gallery_satellite | 951 | 951 | 39| - | Gallery_ground | 2,921 | 793 | 39| - - cameras: None. - - datamanager = torchreid.data.ImageDataManager( - root='reid-data', - sources='university1652', - targets='university1652', - height=256, - width=256, - batch_size_train=32, - batch_size_test=100, - transforms=['random_flip', 'random_crop'] - ) - """ - dataset_dir = 'university1652' - dataset_url = 'https://drive.google.com/uc?id=1iVnP4gjw-iHXa0KerZQ1IfIO0i1jADsR' - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - print(self.dataset_dir) - if not os.path.isdir(self.dataset_dir): - os.mkdir(self.dataset_dir) - gdown.download( - self.dataset_url, self.dataset_dir + 'data.zip', quiet=False - ) - os.system('unzip %s' % (self.dataset_dir + 'data.zip')) - self.train_dir = osp.join( - self.dataset_dir, 'University-Release/train/' - ) - self.query_dir = osp.join( - self.dataset_dir, 'University-Release/test/query_drone' - ) - self.gallery_dir = osp.join( - self.dataset_dir, 'University-Release/test/gallery_satellite' - ) - - required_files = [ - self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir - ] - self.check_before_run(required_files) - - self.fake_camid = 0 - train = self.process_dir(self.train_dir, relabel=True, train=True) - query = self.process_dir(self.query_dir, relabel=False) - gallery = self.process_dir(self.gallery_dir, relabel=False) - - super(University1652, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path, relabel=False, train=False): - IMG_EXTENSIONS = ( - '.jpg', '.jpeg', '.png', '.ppm', '.bmp', '.pgm', '.tif', '.tiff', - '.webp' - ) - if train: - img_paths = glob.glob(osp.join(dir_path, '*/*/*')) - else: - img_paths = glob.glob(osp.join(dir_path, '*/*')) - pid_container = set() - for img_path in img_paths: - if not img_path.lower().endswith(IMG_EXTENSIONS): - continue - pid = int(os.path.basename(os.path.dirname(img_path))) - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - data = [] - # no camera for university - for img_path in img_paths: - if not img_path.lower().endswith(IMG_EXTENSIONS): - continue - pid = int(os.path.basename(os.path.dirname(img_path))) - if relabel: - pid = pid2label[pid] - data.append((img_path, pid, self.fake_camid)) - self.fake_camid += 1 - return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py deleted file mode 100644 index efc64c1460..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py +++ /dev/null @@ -1,175 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import numpy as np -import os.path as osp - -from torchreid.utils import read_json, write_json - -from ..dataset import ImageDataset - - -class VIPeR(ImageDataset): - """VIPeR. - - Reference: - Gray et al. Evaluating appearance models for recognition, reacquisition, and tracking. PETS 2007. - - URL: ``_ - - Dataset statistics: - - identities: 632. - - images: 632 x 2 = 1264. - - cameras: 2. - """ - dataset_dir = 'viper' - dataset_url = 'http://users.soe.ucsc.edu/~manduchi/VIPeR.v1.0.zip' - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.cam_a_dir = osp.join(self.dataset_dir, 'VIPeR', 'cam_a') - self.cam_b_dir = osp.join(self.dataset_dir, 'VIPeR', 'cam_b') - self.split_path = osp.join(self.dataset_dir, 'splits.json') - - required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, ' - 'but expected between 0 and {}'.format( - split_id, - len(splits) - 1 - ) - ) - split = splits[split_id] - - train = split['train'] - query = split['query'] # query and gallery share the same images - gallery = split['gallery'] - - train = [tuple(item) for item in train] - query = [tuple(item) for item in query] - gallery = [tuple(item) for item in gallery] - - super(VIPeR, self).__init__(train, query, gallery, **kwargs) - - def prepare_split(self): - if not osp.exists(self.split_path): - print('Creating 10 random splits of train ids and test ids') - - cam_a_imgs = sorted(glob.glob(osp.join(self.cam_a_dir, '*.bmp'))) - cam_b_imgs = sorted(glob.glob(osp.join(self.cam_b_dir, '*.bmp'))) - assert len(cam_a_imgs) == len(cam_b_imgs) - num_pids = len(cam_a_imgs) - print('Number of identities: {}'.format(num_pids)) - num_train_pids = num_pids // 2 - """ - In total, there will be 20 splits because each random split creates two - sub-splits, one using cameraA as query and cameraB as gallery - while the other using cameraB as query and cameraA as gallery. - Therefore, results should be averaged over 20 splits (split_id=0~19). - - In practice, a model trained on split_id=0 can be applied to split_id=0&1 - as split_id=0&1 share the same training data (so on and so forth). - """ - splits = [] - for _ in range(10): - order = np.arange(num_pids) - np.random.shuffle(order) - train_idxs = order[:num_train_pids] - test_idxs = order[num_train_pids:] - assert not bool(set(train_idxs) & set(test_idxs)), \ - 'Error: train and test overlap' - - train = [] - for pid, idx in enumerate(train_idxs): - cam_a_img = cam_a_imgs[idx] - cam_b_img = cam_b_imgs[idx] - train.append((cam_a_img, pid, 0)) - train.append((cam_b_img, pid, 1)) - - test_a = [] - test_b = [] - for pid, idx in enumerate(test_idxs): - cam_a_img = cam_a_imgs[idx] - cam_b_img = cam_b_imgs[idx] - test_a.append((cam_a_img, pid, 0)) - test_b.append((cam_b_img, pid, 1)) - - # use cameraA as query and cameraB as gallery - split = { - 'train': train, - 'query': test_a, - 'gallery': test_b, - 'num_train_pids': num_train_pids, - 'num_query_pids': num_pids - num_train_pids, - 'num_gallery_pids': num_pids - num_train_pids - } - splits.append(split) - - # use cameraB as query and cameraA as gallery - split = { - 'train': train, - 'query': test_b, - 'gallery': test_a, - 'num_train_pids': num_train_pids, - 'num_query_pids': num_pids - num_train_pids, - 'num_gallery_pids': num_pids - num_train_pids - } - splits.append(split) - - print('Totally {} splits are created'.format(len(splits))) - write_json(splits, self.split_path) - print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py deleted file mode 100644 index 141e7110c1..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py +++ /dev/null @@ -1,53 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .mars import Mars -from .ilidsvid import iLIDSVID -from .prid2011 import PRID2011 -from .dukemtmcvidreid import DukeMTMCVidReID diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py deleted file mode 100644 index 895ff0b38f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py +++ /dev/null @@ -1,175 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import os.path as osp -import warnings - -from torchreid.utils import read_json, write_json - -from ..dataset import VideoDataset - - -class DukeMTMCVidReID(VideoDataset): - """DukeMTMCVidReID. - - Reference: - - Ristani et al. Performance Measures and a Data Set for Multi-Target, - Multi-Camera Tracking. ECCVW 2016. - - Wu et al. Exploit the Unknown Gradually: One-Shot Video-Based Person - Re-Identification by Stepwise Learning. CVPR 2018. - - URL: ``_ - - Dataset statistics: - - identities: 702 (train) + 702 (test). - - tracklets: 2196 (train) + 2636 (test). - """ - dataset_dir = 'dukemtmc-vidreid' - dataset_url = 'http://vision.cs.duke.edu/DukeMTMC/data/misc/DukeMTMC-VideoReID.zip' - - def __init__(self, root='', min_seq_len=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.train_dir = osp.join(self.dataset_dir, 'DukeMTMC-VideoReID/train') - self.query_dir = osp.join(self.dataset_dir, 'DukeMTMC-VideoReID/query') - self.gallery_dir = osp.join( - self.dataset_dir, 'DukeMTMC-VideoReID/gallery' - ) - self.split_train_json_path = osp.join( - self.dataset_dir, 'split_train.json' - ) - self.split_query_json_path = osp.join( - self.dataset_dir, 'split_query.json' - ) - self.split_gallery_json_path = osp.join( - self.dataset_dir, 'split_gallery.json' - ) - self.min_seq_len = min_seq_len - - required_files = [ - self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir - ] - self.check_before_run(required_files) - - train = self.process_dir( - self.train_dir, self.split_train_json_path, relabel=True - ) - query = self.process_dir( - self.query_dir, self.split_query_json_path, relabel=False - ) - gallery = self.process_dir( - self.gallery_dir, self.split_gallery_json_path, relabel=False - ) - - super(DukeMTMCVidReID, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dir_path, json_path, relabel): - if osp.exists(json_path): - split = read_json(json_path) - return split['tracklets'] - - print('=> Generating split json file (** this might take a while **)') - pdirs = glob.glob(osp.join(dir_path, '*')) # avoid .DS_Store - print( - 'Processing "{}" with {} person identities'.format( - dir_path, len(pdirs) - ) - ) - - pid_container = set() - for pdir in pdirs: - pid = int(osp.basename(pdir)) - pid_container.add(pid) - pid2label = {pid: label for label, pid in enumerate(pid_container)} - - tracklets = [] - for pdir in pdirs: - pid = int(osp.basename(pdir)) - if relabel: - pid = pid2label[pid] - tdirs = glob.glob(osp.join(pdir, '*')) - for tdir in tdirs: - raw_img_paths = glob.glob(osp.join(tdir, '*.jpg')) - num_imgs = len(raw_img_paths) - - if num_imgs < self.min_seq_len: - continue - - img_paths = [] - for img_idx in range(num_imgs): - # some tracklet starts from 0002 instead of 0001 - img_idx_name = 'F' + str(img_idx + 1).zfill(4) - res = glob.glob( - osp.join(tdir, '*' + img_idx_name + '*.jpg') - ) - if len(res) == 0: - warnings.warn( - 'Index name {} in {} is missing, skip'.format( - img_idx_name, tdir - ) - ) - continue - img_paths.append(res[0]) - img_name = osp.basename(img_paths[0]) - if img_name.find('_') == -1: - # old naming format: 0001C6F0099X30823.jpg - camid = int(img_name[5]) - 1 - else: - # new naming format: 0001_C6_F0099_X30823.jpg - camid = int(img_name[6]) - 1 - img_paths = tuple(img_paths) - tracklets.append((img_paths, pid, camid)) - - print('Saving split to {}'.format(json_path)) - split_dict = {'tracklets': tracklets} - write_json(split_dict, json_path) - - return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py deleted file mode 100644 index 7e61913d2b..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py +++ /dev/null @@ -1,190 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import os.path as osp -from scipy.io import loadmat - -from torchreid.utils import read_json, write_json - -from ..dataset import VideoDataset - - -class iLIDSVID(VideoDataset): - """iLIDS-VID. - - Reference: - Wang et al. Person Re-Identification by Video Ranking. ECCV 2014. - - URL: ``_ - - Dataset statistics: - - identities: 300. - - tracklets: 600. - - cameras: 2. - """ - dataset_dir = 'ilids-vid' - dataset_url = 'http://www.eecs.qmul.ac.uk/~xiatian/iLIDS-VID/iLIDS-VID.tar' - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.data_dir = osp.join(self.dataset_dir, 'i-LIDS-VID') - self.split_dir = osp.join(self.dataset_dir, 'train-test people splits') - self.split_mat_path = osp.join( - self.split_dir, 'train_test_splits_ilidsvid.mat' - ) - self.split_path = osp.join(self.dataset_dir, 'splits.json') - self.cam_1_path = osp.join( - self.dataset_dir, 'i-LIDS-VID/sequences/cam1' - ) - self.cam_2_path = osp.join( - self.dataset_dir, 'i-LIDS-VID/sequences/cam2' - ) - - required_files = [self.dataset_dir, self.data_dir, self.split_dir] - self.check_before_run(required_files) - - self.prepare_split() - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, but expected between 0 and {}' - .format(split_id, - len(splits) - 1) - ) - split = splits[split_id] - train_dirs, test_dirs = split['train'], split['test'] - - train = self.process_data(train_dirs, cam1=True, cam2=True) - query = self.process_data(test_dirs, cam1=True, cam2=False) - gallery = self.process_data(test_dirs, cam1=False, cam2=True) - - super(iLIDSVID, self).__init__(train, query, gallery, **kwargs) - - def prepare_split(self): - if not osp.exists(self.split_path): - print('Creating splits ...') - mat_split_data = loadmat(self.split_mat_path)['ls_set'] - - num_splits = mat_split_data.shape[0] - num_total_ids = mat_split_data.shape[1] - assert num_splits == 10 - assert num_total_ids == 300 - num_ids_each = num_total_ids // 2 - - # pids in mat_split_data are indices, so we need to transform them - # to real pids - person_cam1_dirs = sorted( - glob.glob(osp.join(self.cam_1_path, '*')) - ) - person_cam2_dirs = sorted( - glob.glob(osp.join(self.cam_2_path, '*')) - ) - - person_cam1_dirs = [ - osp.basename(item) for item in person_cam1_dirs - ] - person_cam2_dirs = [ - osp.basename(item) for item in person_cam2_dirs - ] - - # make sure persons in one camera view can be found in the other camera view - assert set(person_cam1_dirs) == set(person_cam2_dirs) - - splits = [] - for i_split in range(num_splits): - # first 50% for testing and the remaining for training, following Wang et al. ECCV'14. - train_idxs = sorted( - list(mat_split_data[i_split, num_ids_each:]) - ) - test_idxs = sorted( - list(mat_split_data[i_split, :num_ids_each]) - ) - - train_idxs = [int(i) - 1 for i in train_idxs] - test_idxs = [int(i) - 1 for i in test_idxs] - - # transform pids to person dir names - train_dirs = [person_cam1_dirs[i] for i in train_idxs] - test_dirs = [person_cam1_dirs[i] for i in test_idxs] - - split = {'train': train_dirs, 'test': test_dirs} - splits.append(split) - - print( - 'Totally {} splits are created, following Wang et al. ECCV\'14' - .format(len(splits)) - ) - print('Split file is saved to {}'.format(self.split_path)) - write_json(splits, self.split_path) - - def process_data(self, dirnames, cam1=True, cam2=True): - tracklets = [] - dirname2pid = {dirname: i for i, dirname in enumerate(dirnames)} - - for dirname in dirnames: - if cam1: - person_dir = osp.join(self.cam_1_path, dirname) - img_names = glob.glob(osp.join(person_dir, '*.png')) - assert len(img_names) > 0 - img_names = tuple(img_names) - pid = dirname2pid[dirname] - tracklets.append((img_names, pid, 0)) - - if cam2: - person_dir = osp.join(self.cam_2_path, dirname) - img_names = glob.glob(osp.join(person_dir, '*.png')) - assert len(img_names) > 0 - img_names = tuple(img_names) - pid = dirname2pid[dirname] - tracklets.append((img_names, pid, 1)) - - return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py deleted file mode 100644 index fb6d215668..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py +++ /dev/null @@ -1,180 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import os.path as osp -import warnings -from scipy.io import loadmat - -from ..dataset import VideoDataset - - -class Mars(VideoDataset): - """MARS. - - Reference: - Zheng et al. MARS: A Video Benchmark for Large-Scale Person Re-identification. ECCV 2016. - - URL: ``_ - - Dataset statistics: - - identities: 1261. - - tracklets: 8298 (train) + 1980 (query) + 9330 (gallery). - - cameras: 6. - """ - dataset_dir = 'mars' - dataset_url = None - - def __init__(self, root='', **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.train_name_path = osp.join( - self.dataset_dir, 'info/train_name.txt' - ) - self.test_name_path = osp.join(self.dataset_dir, 'info/test_name.txt') - self.track_train_info_path = osp.join( - self.dataset_dir, 'info/tracks_train_info.mat' - ) - self.track_test_info_path = osp.join( - self.dataset_dir, 'info/tracks_test_info.mat' - ) - self.query_IDX_path = osp.join(self.dataset_dir, 'info/query_IDX.mat') - - required_files = [ - self.dataset_dir, self.train_name_path, self.test_name_path, - self.track_train_info_path, self.track_test_info_path, - self.query_IDX_path - ] - self.check_before_run(required_files) - - train_names = self.get_names(self.train_name_path) - test_names = self.get_names(self.test_name_path) - track_train = loadmat(self.track_train_info_path - )['track_train_info'] # numpy.ndarray (8298, 4) - track_test = loadmat(self.track_test_info_path - )['track_test_info'] # numpy.ndarray (12180, 4) - query_IDX = loadmat(self.query_IDX_path - )['query_IDX'].squeeze() # numpy.ndarray (1980,) - query_IDX -= 1 # index from 0 - track_query = track_test[query_IDX, :] - gallery_IDX = [ - i for i in range(track_test.shape[0]) if i not in query_IDX - ] - track_gallery = track_test[gallery_IDX, :] - - train = self.process_data( - train_names, track_train, home_dir='bbox_train', relabel=True - ) - query = self.process_data( - test_names, track_query, home_dir='bbox_test', relabel=False - ) - gallery = self.process_data( - test_names, track_gallery, home_dir='bbox_test', relabel=False - ) - - super(Mars, self).__init__(train, query, gallery, **kwargs) - - def get_names(self, fpath): - names = [] - with open(fpath, 'r') as f: - for line in f: - new_line = line.rstrip() - names.append(new_line) - return names - - def process_data( - self, names, meta_data, home_dir=None, relabel=False, min_seq_len=0 - ): - assert home_dir in ['bbox_train', 'bbox_test'] - num_tracklets = meta_data.shape[0] - pid_list = list(set(meta_data[:, 2].tolist())) - - if relabel: - pid2label = {pid: label for label, pid in enumerate(pid_list)} - tracklets = [] - - for tracklet_idx in range(num_tracklets): - data = meta_data[tracklet_idx, ...] - start_index, end_index, pid, camid = data - if pid == -1: - continue # junk images are just ignored - assert 1 <= camid <= 6 - if relabel: - pid = pid2label[pid] - camid -= 1 # index starts from 0 - img_names = names[start_index - 1:end_index] - - # make sure image names correspond to the same person - pnames = [img_name[:4] for img_name in img_names] - assert len( - set(pnames) - ) == 1, 'Error: a single tracklet contains different person images' - - # make sure all images are captured under the same camera - camnames = [img_name[5] for img_name in img_names] - assert len( - set(camnames) - ) == 1, 'Error: images are captured under different cameras!' - - # append image names with directory information - img_paths = [ - osp.join(self.dataset_dir, home_dir, img_name[:4], img_name) - for img_name in img_names - ] - if len(img_paths) >= min_seq_len: - img_paths = tuple(img_paths) - tracklets.append((img_paths, pid, camid)) - - return tracklets - - def combine_all(self): - warnings.warn( - 'Some query IDs do not appear in gallery. Therefore, combineall ' - 'does not make any difference to Mars' - ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py deleted file mode 100644 index 89598b85a9..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py +++ /dev/null @@ -1,127 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import glob -import os.path as osp - -from torchreid.utils import read_json - -from ..dataset import VideoDataset - - -class PRID2011(VideoDataset): - """PRID2011. - - Reference: - Hirzer et al. Person Re-Identification by Descriptive and - Discriminative Classification. SCIA 2011. - - URL: ``_ - - Dataset statistics: - - identities: 200. - - tracklets: 400. - - cameras: 2. - """ - dataset_dir = 'prid2011' - dataset_url = None - - def __init__(self, root='', split_id=0, **kwargs): - self.root = osp.abspath(osp.expanduser(root)) - self.dataset_dir = osp.join(self.root, self.dataset_dir) - self.download_dataset(self.dataset_dir, self.dataset_url) - - self.split_path = osp.join(self.dataset_dir, 'splits_prid2011.json') - self.cam_a_dir = osp.join( - self.dataset_dir, 'prid_2011', 'multi_shot', 'cam_a' - ) - self.cam_b_dir = osp.join( - self.dataset_dir, 'prid_2011', 'multi_shot', 'cam_b' - ) - - required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] - self.check_before_run(required_files) - - splits = read_json(self.split_path) - if split_id >= len(splits): - raise ValueError( - 'split_id exceeds range, received {}, but expected between 0 and {}' - .format(split_id, - len(splits) - 1) - ) - split = splits[split_id] - train_dirs, test_dirs = split['train'], split['test'] - - train = self.process_dir(train_dirs, cam1=True, cam2=True) - query = self.process_dir(test_dirs, cam1=True, cam2=False) - gallery = self.process_dir(test_dirs, cam1=False, cam2=True) - - super(PRID2011, self).__init__(train, query, gallery, **kwargs) - - def process_dir(self, dirnames, cam1=True, cam2=True): - tracklets = [] - dirname2pid = {dirname: i for i, dirname in enumerate(dirnames)} - - for dirname in dirnames: - if cam1: - person_dir = osp.join(self.cam_a_dir, dirname) - img_names = glob.glob(osp.join(person_dir, '*.png')) - assert len(img_names) > 0 - img_names = tuple(img_names) - pid = dirname2pid[dirname] - tracklets.append((img_names, pid, 0)) - - if cam2: - person_dir = osp.join(self.cam_b_dir, dirname) - img_names = glob.glob(osp.join(person_dir, '*.png')) - assert len(img_names) > 0 - img_names = tuple(img_names) - pid = dirname2pid[dirname] - tracklets.append((img_names, pid, 1)) - - return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py deleted file mode 100644 index daf0d026c3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py +++ /dev/null @@ -1,292 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import copy -import numpy as np -import random -from collections import defaultdict -from torch.utils.data.sampler import Sampler, RandomSampler, SequentialSampler - -AVAI_SAMPLERS = [ - 'RandomIdentitySampler', 'SequentialSampler', 'RandomSampler', - 'RandomDomainSampler', 'RandomDatasetSampler' -] - - -class RandomIdentitySampler(Sampler): - """Randomly samples N identities each with K instances. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - num_instances (int): number of instances per identity in a batch. - """ - - def __init__(self, data_source, batch_size, num_instances): - if batch_size < num_instances: - raise ValueError( - 'batch_size={} must be no less ' - 'than num_instances={}'.format(batch_size, num_instances) - ) - - self.data_source = data_source - self.batch_size = batch_size - self.num_instances = num_instances - self.num_pids_per_batch = self.batch_size // self.num_instances - self.index_dic = defaultdict(list) - for index, items in enumerate(data_source): - pid = items[1] - self.index_dic[pid].append(index) - self.pids = list(self.index_dic.keys()) - assert len(self.pids) >= self.num_pids_per_batch - - # estimate number of examples in an epoch - # TODO: improve precision - self.length = 0 - for pid in self.pids: - idxs = self.index_dic[pid] - num = len(idxs) - if num < self.num_instances: - num = self.num_instances - self.length += num - num % self.num_instances - - def __iter__(self): - batch_idxs_dict = defaultdict(list) - - for pid in self.pids: - idxs = copy.deepcopy(self.index_dic[pid]) - if len(idxs) < self.num_instances: - idxs = np.random.choice( - idxs, size=self.num_instances, replace=True - ) - random.shuffle(idxs) - batch_idxs = [] - for idx in idxs: - batch_idxs.append(idx) - if len(batch_idxs) == self.num_instances: - batch_idxs_dict[pid].append(batch_idxs) - batch_idxs = [] - - avai_pids = copy.deepcopy(self.pids) - final_idxs = [] - - while len(avai_pids) >= self.num_pids_per_batch: - selected_pids = random.sample(avai_pids, self.num_pids_per_batch) - for pid in selected_pids: - batch_idxs = batch_idxs_dict[pid].pop(0) - final_idxs.extend(batch_idxs) - if len(batch_idxs_dict[pid]) == 0: - avai_pids.remove(pid) - - return iter(final_idxs) - - def __len__(self): - return self.length - - -class RandomDomainSampler(Sampler): - """Random domain sampler. - - We consider each camera as a visual domain. - - How does the sampling work: - 1. Randomly sample N cameras (based on the "camid" label). - 2. From each camera, randomly sample K images. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - n_domain (int): number of cameras to sample in a batch. - """ - - def __init__(self, data_source, batch_size, n_domain): - self.data_source = data_source - - # Keep track of image indices for each domain - self.domain_dict = defaultdict(list) - for i, items in enumerate(data_source): - camid = items[2] - self.domain_dict[camid].append(i) - self.domains = list(self.domain_dict.keys()) - - # Make sure each domain can be assigned an equal number of images - if n_domain is None or n_domain <= 0: - n_domain = len(self.domains) - assert batch_size % n_domain == 0 - self.n_img_per_domain = batch_size // n_domain - - self.batch_size = batch_size - self.n_domain = n_domain - self.length = len(list(self.__iter__())) - - def __iter__(self): - domain_dict = copy.deepcopy(self.domain_dict) - final_idxs = [] - stop_sampling = False - - while not stop_sampling: - selected_domains = random.sample(self.domains, self.n_domain) - - for domain in selected_domains: - idxs = domain_dict[domain] - selected_idxs = random.sample(idxs, self.n_img_per_domain) - final_idxs.extend(selected_idxs) - - for idx in selected_idxs: - domain_dict[domain].remove(idx) - - remaining = len(domain_dict[domain]) - if remaining < self.n_img_per_domain: - stop_sampling = True - - return iter(final_idxs) - - def __len__(self): - return self.length - - -class RandomDatasetSampler(Sampler): - """Random dataset sampler. - - How does the sampling work: - 1. Randomly sample N datasets (based on the "dsetid" label). - 2. From each dataset, randomly sample K images. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - n_dataset (int): number of datasets to sample in a batch. - """ - - def __init__(self, data_source, batch_size, n_dataset): - self.data_source = data_source - - # Keep track of image indices for each dataset - self.dataset_dict = defaultdict(list) - for i, items in enumerate(data_source): - dsetid = items[3] - self.dataset_dict[dsetid].append(i) - self.datasets = list(self.dataset_dict.keys()) - - # Make sure each dataset can be assigned an equal number of images - if n_dataset is None or n_dataset <= 0: - n_dataset = len(self.datasets) - assert batch_size % n_dataset == 0 - self.n_img_per_dset = batch_size // n_dataset - - self.batch_size = batch_size - self.n_dataset = n_dataset - self.length = len(list(self.__iter__())) - - def __iter__(self): - dataset_dict = copy.deepcopy(self.dataset_dict) - final_idxs = [] - stop_sampling = False - - while not stop_sampling: - selected_datasets = random.sample(self.datasets, self.n_dataset) - - for dset in selected_datasets: - idxs = dataset_dict[dset] - selected_idxs = random.sample(idxs, self.n_img_per_dset) - final_idxs.extend(selected_idxs) - - for idx in selected_idxs: - dataset_dict[dset].remove(idx) - - remaining = len(dataset_dict[dset]) - if remaining < self.n_img_per_dset: - stop_sampling = True - - return iter(final_idxs) - - def __len__(self): - return self.length - - -def build_train_sampler( - data_source, - train_sampler, - batch_size=32, - num_instances=4, - num_cams=1, - num_datasets=1, - **kwargs -): - """Builds a training sampler. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid). - train_sampler (str): sampler name (default: ``RandomSampler``). - batch_size (int, optional): batch size. Default is 32. - num_instances (int, optional): number of instances per identity in a - batch (when using ``RandomIdentitySampler``). Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - """ - assert train_sampler in AVAI_SAMPLERS, \ - 'train_sampler must be one of {}, but got {}'.format(AVAI_SAMPLERS, train_sampler) - - if train_sampler == 'RandomIdentitySampler': - sampler = RandomIdentitySampler(data_source, batch_size, num_instances) - - elif train_sampler == 'RandomDomainSampler': - sampler = RandomDomainSampler(data_source, batch_size, num_cams) - - elif train_sampler == 'RandomDatasetSampler': - sampler = RandomDatasetSampler(data_source, batch_size, num_datasets) - - elif train_sampler == 'SequentialSampler': - sampler = SequentialSampler(data_source) - - elif train_sampler == 'RandomSampler': - sampler = RandomSampler(data_source) - - return sampler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py deleted file mode 100644 index 3108b81565..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py +++ /dev/null @@ -1,373 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import math -import random -from collections import deque -import torch -from PIL import Image -from torchvision.transforms import ( - Resize, Compose, ToTensor, Normalize, ColorJitter, RandomHorizontalFlip -) - - -class Random2DTranslation(object): - """Randomly translates the input image with a probability. - - Specifically, given a predefined shape (height, width), the input is first - resized with a factor of 1.125, leading to (height*1.125, width*1.125), then - a random crop is performed. Such operation is done with a probability. - - Args: - height (int): target image height. - width (int): target image width. - p (float, optional): probability that this operation takes place. - Default is 0.5. - interpolation (int, optional): desired interpolation. Default is - ``PIL.Image.BILINEAR`` - """ - - def __init__(self, height, width, p=0.5, interpolation=Image.BILINEAR): - self.height = height - self.width = width - self.p = p - self.interpolation = interpolation - - def __call__(self, img): - if random.uniform(0, 1) > self.p: - return img.resize((self.width, self.height), self.interpolation) - - new_width, new_height = int(round(self.width * 1.125) - ), int(round(self.height * 1.125)) - resized_img = img.resize((new_width, new_height), self.interpolation) - x_maxrange = new_width - self.width - y_maxrange = new_height - self.height - x1 = int(round(random.uniform(0, x_maxrange))) - y1 = int(round(random.uniform(0, y_maxrange))) - croped_img = resized_img.crop( - (x1, y1, x1 + self.width, y1 + self.height) - ) - return croped_img - - -class RandomErasing(object): - """Randomly erases an image patch. - - Origin: ``_ - - Reference: - Zhong et al. Random Erasing Data Augmentation. - - Args: - probability (float, optional): probability that this operation takes place. - Default is 0.5. - sl (float, optional): min erasing area. - sh (float, optional): max erasing area. - r1 (float, optional): min aspect ratio. - mean (list, optional): erasing value. - """ - - def __init__( - self, - probability=0.5, - sl=0.02, - sh=0.4, - r1=0.3, - mean=[0.4914, 0.4822, 0.4465] - ): - self.probability = probability - self.mean = mean - self.sl = sl - self.sh = sh - self.r1 = r1 - - def __call__(self, img): - if random.uniform(0, 1) > self.probability: - return img - - for attempt in range(100): - area = img.size()[1] * img.size()[2] - - target_area = random.uniform(self.sl, self.sh) * area - aspect_ratio = random.uniform(self.r1, 1 / self.r1) - - h = int(round(math.sqrt(target_area * aspect_ratio))) - w = int(round(math.sqrt(target_area / aspect_ratio))) - - if w < img.size()[2] and h < img.size()[1]: - x1 = random.randint(0, img.size()[1] - h) - y1 = random.randint(0, img.size()[2] - w) - if img.size()[0] == 3: - img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] - img[1, x1:x1 + h, y1:y1 + w] = self.mean[1] - img[2, x1:x1 + h, y1:y1 + w] = self.mean[2] - else: - img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] - return img - - return img - - -class ColorAugmentation(object): - """Randomly alters the intensities of RGB channels. - - Reference: - Krizhevsky et al. ImageNet Classification with Deep ConvolutionalNeural - Networks. NIPS 2012. - - Args: - p (float, optional): probability that this operation takes place. - Default is 0.5. - """ - - def __init__(self, p=0.5): - self.p = p - self.eig_vec = torch.Tensor( - [ - [0.4009, 0.7192, -0.5675], - [-0.8140, -0.0045, -0.5808], - [0.4203, -0.6948, -0.5836], - ] - ) - self.eig_val = torch.Tensor([[0.2175, 0.0188, 0.0045]]) - - def _check_input(self, tensor): - assert tensor.dim() == 3 and tensor.size(0) == 3 - - def __call__(self, tensor): - if random.uniform(0, 1) > self.p: - return tensor - alpha = torch.normal(mean=torch.zeros_like(self.eig_val)) * 0.1 - quatity = torch.mm(self.eig_val * alpha, self.eig_vec) - tensor = tensor + quatity.view(3, 1, 1) - return tensor - - -class RandomPatch(object): - """Random patch data augmentation. - - There is a patch pool that stores randomly extracted pathces from person images. - - For each input image, RandomPatch - 1) extracts a random patch and stores the patch in the patch pool; - 2) randomly selects a patch from the patch pool and pastes it on the - input (at random position) to simulate occlusion. - - Reference: - - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. - - Zhou et al. Learning Generalisable Omni-Scale Representations - for Person Re-Identification. TPAMI, 2021. - """ - - def __init__( - self, - prob_happen=0.5, - pool_capacity=50000, - min_sample_size=100, - patch_min_area=0.01, - patch_max_area=0.5, - patch_min_ratio=0.1, - prob_rotate=0.5, - prob_flip_leftright=0.5, - ): - self.prob_happen = prob_happen - - self.patch_min_area = patch_min_area - self.patch_max_area = patch_max_area - self.patch_min_ratio = patch_min_ratio - - self.prob_rotate = prob_rotate - self.prob_flip_leftright = prob_flip_leftright - - self.patchpool = deque(maxlen=pool_capacity) - self.min_sample_size = min_sample_size - - def generate_wh(self, W, H): - area = W * H - for attempt in range(100): - target_area = random.uniform( - self.patch_min_area, self.patch_max_area - ) * area - aspect_ratio = random.uniform( - self.patch_min_ratio, 1. / self.patch_min_ratio - ) - h = int(round(math.sqrt(target_area * aspect_ratio))) - w = int(round(math.sqrt(target_area / aspect_ratio))) - if w < W and h < H: - return w, h - return None, None - - def transform_patch(self, patch): - if random.uniform(0, 1) > self.prob_flip_leftright: - patch = patch.transpose(Image.FLIP_LEFT_RIGHT) - if random.uniform(0, 1) > self.prob_rotate: - patch = patch.rotate(random.randint(-10, 10)) - return patch - - def __call__(self, img): - W, H = img.size # original image size - - # collect new patch - w, h = self.generate_wh(W, H) - if w is not None and h is not None: - x1 = random.randint(0, W - w) - y1 = random.randint(0, H - h) - new_patch = img.crop((x1, y1, x1 + w, y1 + h)) - self.patchpool.append(new_patch) - - if len(self.patchpool) < self.min_sample_size: - return img - - if random.uniform(0, 1) > self.prob_happen: - return img - - # paste a randomly selected patch on a random position - patch = random.sample(self.patchpool, 1)[0] - patchW, patchH = patch.size - x1 = random.randint(0, W - patchW) - y1 = random.randint(0, H - patchH) - patch = self.transform_patch(patch) - img.paste(patch, (x1, y1)) - - return img - - -def build_transforms( - height, - width, - transforms='random_flip', - norm_mean=[0.485, 0.456, 0.406], - norm_std=[0.229, 0.224, 0.225], - **kwargs -): - """Builds train and test transform functions. - - Args: - height (int): target image height. - width (int): target image width. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): normalization mean values. Default is ImageNet means. - norm_std (list or None, optional): normalization standard deviation values. Default is - ImageNet standard deviation values. - """ - if transforms is None: - transforms = [] - - if isinstance(transforms, str): - transforms = [transforms] - - if not isinstance(transforms, list): - raise ValueError( - 'transforms must be a list of strings, but found to be {}'.format( - type(transforms) - ) - ) - - if len(transforms) > 0: - transforms = [t.lower() for t in transforms] - - if norm_mean is None or norm_std is None: - norm_mean = [0.485, 0.456, 0.406] # imagenet mean - norm_std = [0.229, 0.224, 0.225] # imagenet std - normalize = Normalize(mean=norm_mean, std=norm_std) - - print('Building train transforms ...') - transform_tr = [] - - print('+ resize to {}x{}'.format(height, width)) - transform_tr += [Resize((height, width))] - - if 'random_flip' in transforms: - print('+ random flip') - transform_tr += [RandomHorizontalFlip()] - - if 'random_crop' in transforms: - print( - '+ random crop (enlarge to {}x{} and ' - 'crop {}x{})'.format( - int(round(height * 1.125)), int(round(width * 1.125)), height, - width - ) - ) - transform_tr += [Random2DTranslation(height, width)] - - if 'random_patch' in transforms: - print('+ random patch') - transform_tr += [RandomPatch()] - - if 'color_jitter' in transforms: - print('+ color jitter') - transform_tr += [ - ColorJitter(brightness=0.2, contrast=0.15, saturation=0, hue=0) - ] - - print('+ to torch tensor of range [0, 1]') - transform_tr += [ToTensor()] - - print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) - transform_tr += [normalize] - - if 'random_erase' in transforms: - print('+ random erase') - transform_tr += [RandomErasing(mean=norm_mean)] - - transform_tr = Compose(transform_tr) - - print('Building test transforms ...') - print('+ resize to {}x{}'.format(height, width)) - print('+ to torch tensor of range [0, 1]') - print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) - - transform_te = Compose([ - Resize((height, width)), - ToTensor(), - normalize, - ]) - - return transform_tr, transform_te diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py deleted file mode 100644 index 7eca9586f7..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py +++ /dev/null @@ -1,52 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .image import ImageSoftmaxEngine, ImageTripletEngine -from .video import VideoSoftmaxEngine, VideoTripletEngine -from .engine import Engine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py deleted file mode 100644 index 4dd250fa95..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py +++ /dev/null @@ -1,547 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import time -import numpy as np -import os.path as osp -import datetime -from collections import OrderedDict -import torch -from torch.nn import functional as F -from torch.utils.tensorboard import SummaryWriter - -from torchreid import metrics -from torchreid.utils import ( - MetricMeter, AverageMeter, re_ranking, open_all_layers, save_checkpoint, - open_specified_layers, visualize_ranked_results -) -from torchreid.losses import DeepSupervision -import os - - -class Engine(object): - r"""A generic base Engine class for both image- and video-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - use_gpu (bool, optional): use gpu. Default is True. - """ - - def __init__(self, datamanager, use_gpu=False, use_npu=False): - self.datamanager = datamanager - self.train_loader = self.datamanager.train_loader - self.test_loader = self.datamanager.test_loader - self.use_gpu = use_gpu - self.use_npu = use_npu - self.writer = None - self.epoch = 0 - - self.model = None - self.optimizer = None - self.scheduler = None - - self._models = OrderedDict() - self._optims = OrderedDict() - self._scheds = OrderedDict() - - def register_model(self, name='model', model=None, optim=None, sched=None): - if self.__dict__.get('_models') is None: - raise AttributeError( - 'Cannot assign model before super().__init__() call' - ) - - if self.__dict__.get('_optims') is None: - raise AttributeError( - 'Cannot assign optim before super().__init__() call' - ) - - if self.__dict__.get('_scheds') is None: - raise AttributeError( - 'Cannot assign sched before super().__init__() call' - ) - - self._models[name] = model - self._optims[name] = optim - self._scheds[name] = sched - - def get_model_names(self, names=None): - names_real = list(self._models.keys()) - if names is not None: - if not isinstance(names, list): - names = [names] - for name in names: - assert name in names_real - return names - else: - return names_real - - def save_model(self, epoch, rank1, save_dir, is_best=False): - names = self.get_model_names() - - for name in names: - save_checkpoint( - { - 'state_dict': self._models[name].state_dict(), - 'epoch': epoch + 1, - 'rank1': rank1, - 'optimizer': self._optims[name].state_dict(), - 'scheduler': self._scheds[name].state_dict() - }, - osp.join(save_dir, name), - is_best=is_best - ) - - def set_model_mode(self, mode='train', names=None): - assert mode in ['train', 'eval', 'test'] - names = self.get_model_names(names) - - for name in names: - if mode == 'train': - self._models[name].train() - else: - self._models[name].eval() - - def get_current_lr(self, names=None): - names = self.get_model_names(names) - name = names[0] - return self._optims[name].param_groups[-1]['lr'] - - def update_lr(self, names=None): - names = self.get_model_names(names) - - for name in names: - if self._scheds[name] is not None: - self._scheds[name].step() - - def run( - self, - save_dir='log', - max_epoch=0, - start_epoch=0, - print_freq=10, - fixbase_epoch=0, - open_layers=None, - start_eval=0, - eval_freq=-1, - test_only=False, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - r"""A unified pipeline for training and evaluating a model. - - Args: - save_dir (str): directory to save model. - max_epoch (int): maximum epoch. - start_epoch (int, optional): starting epoch. Default is 0. - print_freq (int, optional): print_frequency. Default is 10. - fixbase_epoch (int, optional): number of epochs to train ``open_layers`` (new layers) - while keeping base layers frozen. Default is 0. ``fixbase_epoch`` is counted - in ``max_epoch``. - open_layers (str or list, optional): layers (attribute names) open for training. - start_eval (int, optional): from which epoch to start evaluation. Default is 0. - eval_freq (int, optional): evaluation frequency. Default is -1 (meaning evaluation - is only performed at the end of training). - test_only (bool, optional): if True, only runs evaluation on test datasets. - Default is False. - dist_metric (str, optional): distance metric used to compute distance matrix - between query and gallery. Default is "euclidean". - normalize_feature (bool, optional): performs L2 normalization on feature vectors before - computing feature distance. Default is False. - visrank (bool, optional): visualizes ranked results. Default is False. It is recommended to - enable ``visrank`` when ``test_only`` is True. The ranked images will be saved to - "save_dir/visrank_dataset", e.g. "save_dir/visrank_market1501". - visrank_topk (int, optional): top-k ranked images to be visualized. Default is 10. - use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. - Default is False. This should be enabled when using cuhk03 classic split. - ranks (list, optional): cmc ranks to be computed. Default is [1, 5, 10, 20]. - rerank (bool, optional): uses person re-ranking (by Zhong et al. CVPR'17). - Default is False. This is only enabled when test_only=True. - """ - - if visrank and not test_only: - raise ValueError( - 'visrank can be set to True only if test_only=True' - ) - - if test_only: - self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks, - rerank=rerank - ) - return - - if self.writer is None: - self.writer = SummaryWriter(log_dir=save_dir) - - time_start = time.time() - self.start_epoch = start_epoch - self.max_epoch = max_epoch - print('=> Start training') - - device_num = int(os.environ['device_num']) - device_num = 1 if device_num == -1 else device_num - batch_size = int(os.environ['batch_size']) - total_avg = 0.0 - - for self.epoch in range(self.start_epoch, self.max_epoch): - if os.environ['device_num'] != '-1' and os.environ['device_num'] != '1': - self.datamanager.train_sampler.set_epoch(self.epoch) - eve_time = self.train( - print_freq=print_freq, - fixbase_epoch=fixbase_epoch, - open_layers=open_layers - ) - total_avg += eve_time - print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / eve_time, eve_time)) - - if (self.epoch + 1) >= start_eval \ - and eval_freq > 0 \ - and (self.epoch+1) % eval_freq == 0 \ - and (self.epoch + 1) != self.max_epoch: - rank1 = self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks - ) - self.save_model(self.epoch, rank1, save_dir) - - avg_time = total_avg / (self.max_epoch - self.start_epoch) - - if self.max_epoch > 1: - print('=> Final test') - rank1 = self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks - ) - self.save_model(self.epoch, rank1, save_dir) - - elapsed = round(time.time() - time_start) - elapsed = str(datetime.timedelta(seconds=elapsed)) - print('Elapsed {}'.format(elapsed)) - - print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / avg_time, avg_time)) - - if self.writer is not None: - self.writer.close() - - def train(self, print_freq=10, fixbase_epoch=0, open_layers=None): - losses = MetricMeter() - batch_time = AverageMeter() - data_time = AverageMeter() - - self.set_model_mode('train') - - self.two_stepped_transfer_learning( - self.epoch, fixbase_epoch, open_layers - ) - - self.num_batches = len(self.train_loader) - end = time.time() - for self.batch_idx, data in enumerate(self.train_loader): - data_time.update(time.time() - end) - loss_summary = self.forward_backward(data) - batch_time.update(time.time() - end) - losses.update(loss_summary) - - if (self.batch_idx + 1) % print_freq == 0: - nb_this_epoch = self.num_batches - (self.batch_idx + 1) - nb_future_epochs = ( - self.max_epoch - (self.epoch + 1) - ) * self.num_batches - eta_seconds = batch_time.avg * (nb_this_epoch+nb_future_epochs) - eta_str = str(datetime.timedelta(seconds=int(eta_seconds))) - print( - 'epoch: [{0}/{1}][{2}/{3}]\t' - 'time {batch_time.val:.3f} ({batch_time.avg:.3f})\t' - 'data {data_time.val:.3f} ({data_time.avg:.3f})\t' - 'eta {eta}\t' - '{losses}\t' - 'lr {lr:.6f}'.format( - self.epoch + 1, - self.max_epoch, - self.batch_idx + 1, - self.num_batches, - batch_time=batch_time, - data_time=data_time, - eta=eta_str, - losses=losses, - lr=self.get_current_lr() - ) - ) - - if self.writer is not None: - n_iter = self.epoch * self.num_batches + self.batch_idx - self.writer.add_scalar('Train/time', batch_time.avg, n_iter) - self.writer.add_scalar('Train/data', data_time.avg, n_iter) - for name, meter in losses.meters.items(): - self.writer.add_scalar('Train/' + name, meter.avg, n_iter) - self.writer.add_scalar( - 'Train/lr', self.get_current_lr(), n_iter - ) - - end = time.time() - - if self.batch_idx == 4: # 前5个step不记录时间 - batch_time.reset() - - self.update_lr() - return batch_time.avg - - def forward_backward(self, data): - raise NotImplementedError - - def test( - self, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - save_dir='', - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - r"""Tests model on target datasets. - - .. note:: - - This function has been called in ``run()``. - - .. note:: - - The test pipeline implemented in this function suits both image- and - video-reid. In general, a subclass of Engine only needs to re-implement - ``extract_features()`` and ``parse_data_for_eval()`` (most of the time), - but not a must. Please refer to the source code for more details. - """ - self.set_model_mode('eval') - targets = list(self.test_loader.keys()) - - for name in targets: - domain = 'source' if name in self.datamanager.sources else 'target' - print('##### Evaluating {} ({}) #####'.format(name, domain)) - query_loader = self.test_loader[name]['query'] - gallery_loader = self.test_loader[name]['gallery'] - rank1, mAP = self._evaluate( - dataset_name=name, - query_loader=query_loader, - gallery_loader=gallery_loader, - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks, - rerank=rerank - ) - - if self.writer is not None: - self.writer.add_scalar(f'Test/{name}/rank1', rank1, self.epoch) - self.writer.add_scalar(f'Test/{name}/mAP', mAP, self.epoch) - - return rank1 - - @torch.no_grad() - def _evaluate( - self, - dataset_name='', - query_loader=None, - gallery_loader=None, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - save_dir='', - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - batch_time = AverageMeter() - - def _feature_extraction(data_loader): - f_, pids_, camids_ = [], [], [] - for batch_idx, data in enumerate(data_loader): - imgs, pids, camids = self.parse_data_for_eval(data) - if self.use_gpu: - imgs = imgs.cuda() - elif self.use_npu: - imgs = imgs.npu() - end = time.time() - features = self.extract_features(imgs) - batch_time.update(time.time() - end) - features = features.cpu().clone() - f_.append(features) - pids_.extend(pids) - camids_.extend(camids) - f_ = torch.cat(f_, 0) - pids_ = np.asarray(pids_) - camids_ = np.asarray(camids_) - return f_, pids_, camids_ - - print('Extracting features from query set ...') - qf, q_pids, q_camids = _feature_extraction(query_loader) - print('Done, obtained {}-by-{} matrix'.format(qf.size(0), qf.size(1))) - - print('Extracting features from gallery set ...') - gf, g_pids, g_camids = _feature_extraction(gallery_loader) - print('Done, obtained {}-by-{} matrix'.format(gf.size(0), gf.size(1))) - - print('Speed: {:.4f} sec/batch'.format(batch_time.avg)) - - if normalize_feature: - print('Normalzing features with L2 norm ...') - qf = F.normalize(qf, p=2, dim=1) - gf = F.normalize(gf, p=2, dim=1) - - print( - 'Computing distance matrix with metric={} ...'.format(dist_metric) - ) - distmat = metrics.compute_distance_matrix(qf, gf, dist_metric) - distmat = distmat.numpy() - - if rerank: - print('Applying person re-ranking ...') - distmat_qq = metrics.compute_distance_matrix(qf, qf, dist_metric) - distmat_gg = metrics.compute_distance_matrix(gf, gf, dist_metric) - distmat = re_ranking(distmat, distmat_qq, distmat_gg) - - print('Computing CMC and mAP ...') - cmc, mAP = metrics.evaluate_rank( - distmat, - q_pids, - g_pids, - q_camids, - g_camids, - use_metric_cuhk03=use_metric_cuhk03 - ) - - print('** Results **') - print('mAP: {:.1%}'.format(mAP)) - print('CMC curve') - for r in ranks: - print('Rank-{:<3}: {:.1%}'.format(r, cmc[r - 1])) - - if visrank: - visualize_ranked_results( - distmat, - self.datamanager.fetch_test_loaders(dataset_name), - self.datamanager.data_type, - width=self.datamanager.width, - height=self.datamanager.height, - save_dir=osp.join(save_dir, 'visrank_' + dataset_name), - topk=visrank_topk - ) - - return cmc[0], mAP - - def compute_loss(self, criterion, outputs, targets): - if isinstance(outputs, (tuple, list)): - loss = DeepSupervision(criterion, outputs, targets) - else: - loss = criterion(outputs, targets) - return loss - - def extract_features(self, input): - return self.model(input) - - def parse_data_for_train(self, data): - imgs = data['img'] - pids = data['pid'] - return imgs, pids - - def parse_data_for_eval(self, data): - imgs = data['img'] - pids = data['pid'] - camids = data['camid'] - return imgs, pids, camids - - def two_stepped_transfer_learning( - self, epoch, fixbase_epoch, open_layers, model=None - ): - """Two-stepped transfer learning. - - The idea is to freeze base layers for a certain number of epochs - and then open all layers for training. - - Reference: https://arxiv.org/abs/1611.05244 - """ - model = self.model if model is None else model - if model is None: - return - - if (epoch + 1) <= fixbase_epoch and open_layers is not None: - print( - '* Only train {} (epoch: {}/{})'.format( - open_layers, epoch + 1, fixbase_epoch - ) - ) - open_specified_layers(model, open_layers) - else: - open_all_layers(model) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py deleted file mode 100644 index fef3e0a2e3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py +++ /dev/null @@ -1,51 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .softmax import ImageSoftmaxEngine -from .triplet import ImageTripletEngine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py deleted file mode 100644 index 0a836672d5..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py +++ /dev/null @@ -1,157 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - -from torchreid import metrics -from torchreid.losses import CrossEntropyLoss - -from ..engine import Engine - -from apex import amp - -class ImageSoftmaxEngine(Engine): - r"""Softmax-loss engine for image-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - model (nn.Module): model instance. - optimizer (Optimizer): an Optimizer. - scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. - use_gpu (bool, optional): use gpu. Default is True. - label_smooth (bool, optional): use label smoothing regularizer. Default is True. - - Examples:: - - import torchreid - datamanager = torchreid.data.ImageDataManager( - root='path/to/reid-data', - sources='market1501', - height=256, - width=128, - combineall=False, - batch_size=32 - ) - model = torchreid.models.build_model( - name='resnet50', - num_classes=datamanager.num_train_pids, - loss='softmax' - ) - model = model.cuda() - optimizer = torchreid.optim.build_optimizer( - model, optim='adam', lr=0.0003 - ) - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler='single_step', - stepsize=20 - ) - engine = torchreid.engine.ImageSoftmaxEngine( - datamanager, model, optimizer, scheduler=scheduler - ) - engine.run( - max_epoch=60, - save_dir='log/resnet50-softmax-market1501', - print_freq=10 - ) - """ - - def __init__( - self, - datamanager, - model, - optimizer, - scheduler=None, - use_gpu=False, - use_npu=False, - label_smooth=True, - use_amp=False - ): - super(ImageSoftmaxEngine, self).__init__(datamanager, use_gpu, use_npu) - - self.model = model - self.optimizer = optimizer - self.scheduler = scheduler - self.register_model('model', model, optimizer, scheduler) - self.use_amp = use_amp - - - self.criterion = CrossEntropyLoss( - num_classes=self.datamanager.num_train_pids, - use_gpu=self.use_gpu, - use_npu=self.use_npu, - label_smooth=label_smooth - ) - - def forward_backward(self, data): - imgs, pids = self.parse_data_for_train(data) - - if self.use_gpu: - imgs = imgs.cuda() - pids = pids.cuda() - elif self.use_npu: - imgs = imgs.npu() - pids = pids.npu() - - outputs = self.model(imgs) - loss = self.compute_loss(self.criterion, outputs, pids) - - self.optimizer.zero_grad() - if self.use_amp: - with amp.scale_loss(loss, self.optimizer) as scaled_loss: - scaled_loss.backward() - else: - loss.backward() - self.optimizer.step() - - loss_summary = { - 'loss': loss.item(), - 'acc': metrics.accuracy(outputs, pids)[0].item() - } - - return loss_summary diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py deleted file mode 100644 index 64a807734b..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py +++ /dev/null @@ -1,169 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - -from torchreid import metrics -from torchreid.losses import TripletLoss, CrossEntropyLoss - -from ..engine import Engine - - -class ImageTripletEngine(Engine): - r"""Triplet-loss engine for image-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - model (nn.Module): model instance. - optimizer (Optimizer): an Optimizer. - margin (float, optional): margin for triplet loss. Default is 0.3. - weight_t (float, optional): weight for triplet loss. Default is 1. - weight_x (float, optional): weight for softmax loss. Default is 1. - scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. - use_gpu (bool, optional): use gpu. Default is True. - label_smooth (bool, optional): use label smoothing regularizer. Default is True. - - Examples:: - - import torchreid - datamanager = torchreid.data.ImageDataManager( - root='path/to/reid-data', - sources='market1501', - height=256, - width=128, - combineall=False, - batch_size=32, - num_instances=4, - train_sampler='RandomIdentitySampler' # this is important - ) - model = torchreid.models.build_model( - name='resnet50', - num_classes=datamanager.num_train_pids, - loss='triplet' - ) - model = model.cuda() - optimizer = torchreid.optim.build_optimizer( - model, optim='adam', lr=0.0003 - ) - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler='single_step', - stepsize=20 - ) - engine = torchreid.engine.ImageTripletEngine( - datamanager, model, optimizer, margin=0.3, - weight_t=0.7, weight_x=1, scheduler=scheduler - ) - engine.run( - max_epoch=60, - save_dir='log/resnet50-triplet-market1501', - print_freq=10 - ) - """ - - def __init__( - self, - datamanager, - model, - optimizer, - margin=0.3, - weight_t=1, - weight_x=1, - scheduler=None, - use_gpu=True, - label_smooth=True - ): - super(ImageTripletEngine, self).__init__(datamanager, use_gpu) - - self.model = model - self.optimizer = optimizer - self.scheduler = scheduler - self.register_model('model', model, optimizer, scheduler) - - assert weight_t >= 0 and weight_x >= 0 - assert weight_t + weight_x > 0 - self.weight_t = weight_t - self.weight_x = weight_x - - self.criterion_t = TripletLoss(margin=margin) - self.criterion_x = CrossEntropyLoss( - num_classes=self.datamanager.num_train_pids, - use_gpu=self.use_gpu, - label_smooth=label_smooth - ) - - def forward_backward(self, data): - imgs, pids = self.parse_data_for_train(data) - - if self.use_gpu: - imgs = imgs.cuda() - pids = pids.cuda() - - outputs, features = self.model(imgs) - - loss = 0 - loss_summary = {} - - if self.weight_t > 0: - loss_t = self.compute_loss(self.criterion_t, features, pids) - loss += self.weight_t * loss_t - loss_summary['loss_t'] = loss_t.item() - - if self.weight_x > 0: - loss_x = self.compute_loss(self.criterion_x, outputs, pids) - loss += self.weight_x * loss_x - loss_summary['loss_x'] = loss_x.item() - loss_summary['acc'] = metrics.accuracy(outputs, pids)[0].item() - - assert loss_summary - - self.optimizer.zero_grad() - loss.backward() - self.optimizer.step() - - return loss_summary diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py deleted file mode 100644 index 28fc05991d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py +++ /dev/null @@ -1,51 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .softmax import VideoSoftmaxEngine -from .triplet import VideoTripletEngine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py deleted file mode 100644 index 44c42bb795..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py +++ /dev/null @@ -1,156 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch - -from torchreid.engine.image import ImageSoftmaxEngine - - -class VideoSoftmaxEngine(ImageSoftmaxEngine): - """Softmax-loss engine for video-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - model (nn.Module): model instance. - optimizer (Optimizer): an Optimizer. - scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. - use_gpu (bool, optional): use gpu. Default is True. - label_smooth (bool, optional): use label smoothing regularizer. Default is True. - pooling_method (str, optional): how to pool features for a tracklet. - Default is "avg" (average). Choices are ["avg", "max"]. - - Examples:: - - import torch - import torchreid - # Each batch contains batch_size*seq_len images - datamanager = torchreid.data.VideoDataManager( - root='path/to/reid-data', - sources='mars', - height=256, - width=128, - combineall=False, - batch_size=8, # number of tracklets - seq_len=15 # number of images in each tracklet - ) - model = torchreid.models.build_model( - name='resnet50', - num_classes=datamanager.num_train_pids, - loss='softmax' - ) - model = model.cuda() - optimizer = torchreid.optim.build_optimizer( - model, optim='adam', lr=0.0003 - ) - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler='single_step', - stepsize=20 - ) - engine = torchreid.engine.VideoSoftmaxEngine( - datamanager, model, optimizer, scheduler=scheduler, - pooling_method='avg' - ) - engine.run( - max_epoch=60, - save_dir='log/resnet50-softmax-mars', - print_freq=10 - ) - """ - - def __init__( - self, - datamanager, - model, - optimizer, - scheduler=None, - use_gpu=True, - label_smooth=True, - pooling_method='avg' - ): - super(VideoSoftmaxEngine, self).__init__( - datamanager, - model, - optimizer, - scheduler=scheduler, - use_gpu=use_gpu, - label_smooth=label_smooth - ) - self.pooling_method = pooling_method - - def parse_data_for_train(self, data): - imgs = data['img'] - pids = data['pid'] - if imgs.dim() == 5: - # b: batch size - # s: sqeuence length - # c: channel depth - # h: height - # w: width - b, s, c, h, w = imgs.size() - imgs = imgs.view(b * s, c, h, w) - pids = pids.view(b, 1).expand(b, s) - pids = pids.contiguous().view(b * s) - return imgs, pids - - def extract_features(self, input): - # b: batch size - # s: sqeuence length - # c: channel depth - # h: height - # w: width - b, s, c, h, w = input.size() - input = input.view(b * s, c, h, w) - features = self.model(input) - features = features.view(b, s, -1) - if self.pooling_method == 'avg': - features = torch.mean(features, 1) - else: - features = torch.max(features, 1)[0] - return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py deleted file mode 100644 index 5604715d4a..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py +++ /dev/null @@ -1,169 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch - -from torchreid.engine.image import ImageTripletEngine - - -class VideoTripletEngine(ImageTripletEngine): - """Triplet-loss engine for video-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - model (nn.Module): model instance. - optimizer (Optimizer): an Optimizer. - margin (float, optional): margin for triplet loss. Default is 0.3. - weight_t (float, optional): weight for triplet loss. Default is 1. - weight_x (float, optional): weight for softmax loss. Default is 1. - scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. - use_gpu (bool, optional): use gpu. Default is True. - label_smooth (bool, optional): use label smoothing regularizer. Default is True. - pooling_method (str, optional): how to pool features for a tracklet. - Default is "avg" (average). Choices are ["avg", "max"]. - - Examples:: - - import torch - import torchreid - # Each batch contains batch_size*seq_len images - # Each identity is sampled with num_instances tracklets - datamanager = torchreid.data.VideoDataManager( - root='path/to/reid-data', - sources='mars', - height=256, - width=128, - combineall=False, - num_instances=4, - train_sampler='RandomIdentitySampler' - batch_size=8, # number of tracklets - seq_len=15 # number of images in each tracklet - ) - model = torchreid.models.build_model( - name='resnet50', - num_classes=datamanager.num_train_pids, - loss='triplet' - ) - model = model.cuda() - optimizer = torchreid.optim.build_optimizer( - model, optim='adam', lr=0.0003 - ) - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler='single_step', - stepsize=20 - ) - engine = torchreid.engine.VideoTripletEngine( - datamanager, model, optimizer, margin=0.3, - weight_t=0.7, weight_x=1, scheduler=scheduler, - pooling_method='avg' - ) - engine.run( - max_epoch=60, - save_dir='log/resnet50-triplet-mars', - print_freq=10 - ) - """ - - def __init__( - self, - datamanager, - model, - optimizer, - margin=0.3, - weight_t=1, - weight_x=1, - scheduler=None, - use_gpu=True, - label_smooth=True, - pooling_method='avg' - ): - super(VideoTripletEngine, self).__init__( - datamanager, - model, - optimizer, - margin=margin, - weight_t=weight_t, - weight_x=weight_x, - scheduler=scheduler, - use_gpu=use_gpu, - label_smooth=label_smooth - ) - self.pooling_method = pooling_method - - def parse_data_for_train(self, data): - imgs = data['img'] - pids = data['pid'] - if imgs.dim() == 5: - # b: batch size - # s: sqeuence length - # c: channel depth - # h: height - # w: width - b, s, c, h, w = imgs.size() - imgs = imgs.view(b * s, c, h, w) - pids = pids.view(b, 1).expand(b, s) - pids = pids.contiguous().view(b * s) - return imgs, pids - - def extract_features(self, input): - # b: batch size - # s: sqeuence length - # c: channel depth - # h: height - # w: width - b, s, c, h, w = input.size() - input = input.view(b * s, c, h, w) - features = self.model(input) - features = features.view(b, s, -1) - if self.pooling_method == 'avg': - features = torch.mean(features, 1) - else: - features = torch.max(features, 1)[0] - return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py deleted file mode 100644 index 0376bc80de..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py +++ /dev/null @@ -1,68 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - -from .cross_entropy_loss import CrossEntropyLoss -from .hard_mine_triplet_loss import TripletLoss - - -def DeepSupervision(criterion, xs, y): - """DeepSupervision - - Applies criterion to each element in a list. - - Args: - criterion: loss function - xs: tuple of inputs - y: ground truth - """ - loss = 0. - for x in xs: - loss += criterion(x, y) - loss /= len(xs) - return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py deleted file mode 100644 index d043691331..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py +++ /dev/null @@ -1,100 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn - - -class CrossEntropyLoss(nn.Module): - r"""Cross entropy loss with label smoothing regularizer. - - Reference: - Szegedy et al. Rethinking the Inception Architecture for Computer Vision. CVPR 2016. - - With label smoothing, the label :math:`y` for a class is computed by - - .. math:: - \begin{equation} - (1 - \eps) \times y + \frac{\eps}{K}, - \end{equation} - - where :math:`K` denotes the number of classes and :math:`\eps` is a weight. When - :math:`\eps = 0`, the loss function reduces to the normal cross entropy. - - Args: - num_classes (int): number of classes. - eps (float, optional): weight. Default is 0.1. - use_gpu (bool, optional): whether to use gpu devices. Default is True. - label_smooth (bool, optional): whether to apply label smoothing. Default is True. - """ - - def __init__(self, num_classes, eps=0.1, use_gpu=False, use_npu=False, label_smooth=True): - super(CrossEntropyLoss, self).__init__() - self.num_classes = num_classes - self.eps = eps if label_smooth else 0 - self.use_gpu = use_gpu - self.use_npu = use_npu - self.logsoftmax = nn.LogSoftmax(dim=1) - - def forward(self, inputs, targets): - """ - Args: - inputs (torch.Tensor): prediction matrix (before softmax) with - shape (batch_size, num_classes). - targets (torch.LongTensor): ground truth labels with shape (batch_size). - Each position contains the label index. - """ - log_probs = self.logsoftmax(inputs) - zeros = torch.zeros(log_probs.size()) - targets = zeros.scatter_(1, targets.unsqueeze(1).data.cpu(), 1) - if self.use_gpu: - targets = targets.cuda() - elif self.use_npu: - targets = targets.npu() - targets = (1 - self.eps) * targets + self.eps / self.num_classes - return (-targets * log_probs).mean(0).sum() diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py deleted file mode 100644 index ac2d927c92..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py +++ /dev/null @@ -1,95 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn - - -class TripletLoss(nn.Module): - """Triplet loss with hard positive/negative mining. - - Reference: - Hermans et al. In Defense of the Triplet Loss for Person Re-Identification. arXiv:1703.07737. - - Imported from ``_. - - Args: - margin (float, optional): margin for triplet. Default is 0.3. - """ - - def __init__(self, margin=0.3): - super(TripletLoss, self).__init__() - self.margin = margin - self.ranking_loss = nn.MarginRankingLoss(margin=margin) - - def forward(self, inputs, targets): - """ - Args: - inputs (torch.Tensor): feature matrix with shape (batch_size, feat_dim). - targets (torch.LongTensor): ground truth labels with shape (num_classes). - """ - n = inputs.size(0) - - # Compute pairwise distance, replace by the official when merged - dist = torch.pow(inputs, 2).sum(dim=1, keepdim=True).expand(n, n) - dist = dist + dist.t() - dist.addmm_(inputs, inputs.t(), beta=1, alpha=-2) - dist = dist.clamp(min=1e-12).sqrt() # for numerical stability - - # For each anchor, find the hardest positive and negative - mask = targets.expand(n, n).eq(targets.expand(n, n).t()) - dist_ap, dist_an = [], [] - for i in range(n): - dist_ap.append(dist[i][mask[i]].max().unsqueeze(0)) - dist_an.append(dist[i][mask[i] == 0].min().unsqueeze(0)) - dist_ap = torch.cat(dist_ap) - dist_an = torch.cat(dist_an) - - # Compute ranking hinge loss - y = torch.ones_like(dist_an) - return self.ranking_loss(dist_an, dist_ap, y) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py deleted file mode 100644 index b1c17830fa..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py +++ /dev/null @@ -1,52 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .rank import evaluate_rank -from .accuracy import accuracy -from .distance import compute_distance_matrix diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py deleted file mode 100644 index c5145818b8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py +++ /dev/null @@ -1,84 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - - -def accuracy(output, target, topk=(1, )): - """Computes the accuracy over the k top predictions for - the specified values of k. - - Args: - output (torch.Tensor): prediction matrix with shape (batch_size, num_classes). - target (torch.LongTensor): ground truth labels with shape (batch_size). - topk (tuple, optional): accuracy at top-k will be computed. For example, - topk=(1, 5) means accuracy at top-1 and top-5 will be computed. - - Returns: - list: accuracy at top-k. - - Examples:: - >>> from torchreid import metrics - >>> metrics.accuracy(output, target) - """ - maxk = max(topk) - batch_size = target.size(0) - - if isinstance(output, (tuple, list)): - output = output[0] - - _, pred = output.topk(maxk, 1, True, True) - pred = pred.t() - correct = pred.eq(target.view(1, -1).expand_as(pred)) - - res = [] - for k in topk: - correct_k = correct[:k].view(-1).float().sum(0, keepdim=True) - acc = correct_k.mul_(100.0 / batch_size) - res.append(acc) - - return res diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py deleted file mode 100644 index f209c03fa2..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py +++ /dev/null @@ -1,127 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch -from torch.nn import functional as F - - -def compute_distance_matrix(input1, input2, metric='euclidean'): - """A wrapper function for computing distance matrix. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - metric (str, optional): "euclidean" or "cosine". - Default is "euclidean". - - Returns: - torch.Tensor: distance matrix. - - Examples:: - >>> from torchreid import metrics - >>> input1 = torch.rand(10, 2048) - >>> input2 = torch.rand(100, 2048) - >>> distmat = metrics.compute_distance_matrix(input1, input2) - >>> distmat.size() # (10, 100) - """ - # check input - assert isinstance(input1, torch.Tensor) - assert isinstance(input2, torch.Tensor) - assert input1.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( - input1.dim() - ) - assert input2.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( - input2.dim() - ) - assert input1.size(1) == input2.size(1) - - if metric == 'euclidean': - distmat = euclidean_squared_distance(input1, input2) - elif metric == 'cosine': - distmat = cosine_distance(input1, input2) - else: - raise ValueError( - 'Unknown distance metric: {}. ' - 'Please choose either "euclidean" or "cosine"'.format(metric) - ) - - return distmat - - -def euclidean_squared_distance(input1, input2): - """Computes euclidean squared distance. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - - Returns: - torch.Tensor: distance matrix. - """ - m, n = input1.size(0), input2.size(0) - mat1 = torch.pow(input1, 2).sum(dim=1, keepdim=True).expand(m, n) - mat2 = torch.pow(input2, 2).sum(dim=1, keepdim=True).expand(n, m).t() - distmat = mat1 + mat2 - distmat.addmm_(input1, input2.t(), beta=1, alpha=-2) - return distmat - - -def cosine_distance(input1, input2): - """Computes cosine distance. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - - Returns: - torch.Tensor: distance matrix. - """ - input1_normed = F.normalize(input1, p=2, dim=1) - input2_normed = F.normalize(input2, p=2, dim=1) - distmat = 1 - torch.mm(input1_normed, input2_normed.t()) - return distmat diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py deleted file mode 100644 index 8e9fc70253..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py +++ /dev/null @@ -1,254 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import numpy as np -import warnings -from collections import defaultdict - -try: - from torchreid.metrics.rank_cylib.rank_cy import evaluate_cy - IS_CYTHON_AVAI = True -except ImportError: - IS_CYTHON_AVAI = False - warnings.warn( - 'Cython evaluation (very fast so highly recommended) is ' - 'unavailable, now use python evaluation.' - ) - - -def eval_cuhk03(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): - """Evaluation with cuhk03 metric - Key: one image for each gallery identity is randomly sampled for each query identity. - Random sampling is performed num_repeats times. - """ - num_repeats = 10 - num_q, num_g = distmat.shape - - if num_g < max_rank: - max_rank = num_g - print( - 'Note: number of gallery samples is quite small, got {}'. - format(num_g) - ) - - indices = np.argsort(distmat, axis=1) - matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) - - # compute cmc curve for each query - all_cmc = [] - all_AP = [] - num_valid_q = 0. # number of valid query - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - order = indices[q_idx] - remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) - keep = np.invert(remove) - - # compute cmc curve - raw_cmc = matches[q_idx][ - keep] # binary vector, positions with value 1 are correct matches - if not np.any(raw_cmc): - # this condition is true when query identity does not appear in gallery - continue - - kept_g_pids = g_pids[order][keep] - g_pids_dict = defaultdict(list) - for idx, pid in enumerate(kept_g_pids): - g_pids_dict[pid].append(idx) - - cmc = 0. - for repeat_idx in range(num_repeats): - mask = np.zeros(len(raw_cmc), dtype=np.bool) - for _, idxs in g_pids_dict.items(): - # randomly sample one image for each gallery person - rnd_idx = np.random.choice(idxs) - mask[rnd_idx] = True - masked_raw_cmc = raw_cmc[mask] - _cmc = masked_raw_cmc.cumsum() - _cmc[_cmc > 1] = 1 - cmc += _cmc[:max_rank].astype(np.float32) - - cmc /= num_repeats - all_cmc.append(cmc) - # compute AP - num_rel = raw_cmc.sum() - tmp_cmc = raw_cmc.cumsum() - tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] - tmp_cmc = np.asarray(tmp_cmc) * raw_cmc - AP = tmp_cmc.sum() / num_rel - all_AP.append(AP) - num_valid_q += 1. - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - all_cmc = np.asarray(all_cmc).astype(np.float32) - all_cmc = all_cmc.sum(0) / num_valid_q - mAP = np.mean(all_AP) - - return all_cmc, mAP - - -def eval_market1501(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): - """Evaluation with market1501 metric - Key: for each query identity, its gallery images from the same camera view are discarded. - """ - num_q, num_g = distmat.shape - - if num_g < max_rank: - max_rank = num_g - print( - 'Note: number of gallery samples is quite small, got {}'. - format(num_g) - ) - - indices = np.argsort(distmat, axis=1) - matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) - - # compute cmc curve for each query - all_cmc = [] - all_AP = [] - num_valid_q = 0. # number of valid query - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - order = indices[q_idx] - remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) - keep = np.invert(remove) - - # compute cmc curve - raw_cmc = matches[q_idx][ - keep] # binary vector, positions with value 1 are correct matches - if not np.any(raw_cmc): - # this condition is true when query identity does not appear in gallery - continue - - cmc = raw_cmc.cumsum() - cmc[cmc > 1] = 1 - - all_cmc.append(cmc[:max_rank]) - num_valid_q += 1. - - # compute average precision - # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision - num_rel = raw_cmc.sum() - tmp_cmc = raw_cmc.cumsum() - tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] - tmp_cmc = np.asarray(tmp_cmc) * raw_cmc - AP = tmp_cmc.sum() / num_rel - all_AP.append(AP) - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - all_cmc = np.asarray(all_cmc).astype(np.float32) - all_cmc = all_cmc.sum(0) / num_valid_q - mAP = np.mean(all_AP) - - return all_cmc, mAP - - -def evaluate_py( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03 -): - if use_metric_cuhk03: - return eval_cuhk03( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank - ) - else: - return eval_market1501( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank - ) - - -def evaluate_rank( - distmat, - q_pids, - g_pids, - q_camids, - g_camids, - max_rank=50, - use_metric_cuhk03=False, - use_cython=True -): - """Evaluates CMC rank. - - Args: - distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). - q_pids (numpy.ndarray): 1-D array containing person identities - of each query instance. - g_pids (numpy.ndarray): 1-D array containing person identities - of each gallery instance. - q_camids (numpy.ndarray): 1-D array containing camera views under - which each query instance is captured. - g_camids (numpy.ndarray): 1-D array containing camera views under - which each gallery instance is captured. - max_rank (int, optional): maximum CMC rank to be computed. Default is 50. - use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. - Default is False. This should be enabled when using cuhk03 classic split. - use_cython (bool, optional): use cython code for evaluation. Default is True. - This is highly recommended as the cython code can speed up the cmc computation - by more than 10x. This requires Cython to be installed. - """ - if use_cython and IS_CYTHON_AVAI: - return evaluate_cy( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, - use_metric_cuhk03 - ) - else: - return evaluate_py( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, - use_metric_cuhk03 - ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile deleted file mode 100644 index d49e655f85..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile +++ /dev/null @@ -1,6 +0,0 @@ -all: - $(PYTHON) setup.py build_ext --inplace - rm -rf build -clean: - rm -rf build - rm -f rank_cy.c *.so \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py deleted file mode 100644 index ec62d364d8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py +++ /dev/null @@ -1,47 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx deleted file mode 100644 index b4a8690e57..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx +++ /dev/null @@ -1,251 +0,0 @@ -# cython: boundscheck=False, wraparound=False, nonecheck=False, cdivision=True - -from __future__ import print_function -import numpy as np -from libc.stdint cimport int64_t, uint64_t - -import cython - -cimport numpy as np - -import random -from collections import defaultdict - -""" -Compiler directives: -https://github.com/cython/cython/wiki/enhancements-compilerdirectives - -Cython tutorial: -https://cython.readthedocs.io/en/latest/src/userguide/numpy_tutorial.html - -Credit to https://github.com/luzai -""" - - -# Main interface -cpdef evaluate_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=False): - distmat = np.asarray(distmat, dtype=np.float32) - q_pids = np.asarray(q_pids, dtype=np.int64) - g_pids = np.asarray(g_pids, dtype=np.int64) - q_camids = np.asarray(q_camids, dtype=np.int64) - g_camids = np.asarray(g_camids, dtype=np.int64) - if use_metric_cuhk03: - return eval_cuhk03_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank) - return eval_market1501_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank) - - -cpdef eval_cuhk03_cy(float[:,:] distmat, int64_t[:] q_pids, int64_t[:]g_pids, - int64_t[:]q_camids, int64_t[:]g_camids, int64_t max_rank): - - cdef int64_t num_q = distmat.shape[0] - cdef int64_t num_g = distmat.shape[1] - - if num_g < max_rank: - max_rank = num_g - print('Note: number of gallery samples is quite small, got {}'.format(num_g)) - - cdef: - int64_t num_repeats = 10 - int64_t[:,:] indices = np.argsort(distmat, axis=1) - int64_t[:,:] matches = (np.asarray(g_pids)[np.asarray(indices)] == np.asarray(q_pids)[:, np.newaxis]).astype(np.int64) - - float[:,:] all_cmc = np.zeros((num_q, max_rank), dtype=np.float32) - float[:] all_AP = np.zeros(num_q, dtype=np.float32) - float num_valid_q = 0. # number of valid query - - int64_t q_idx, q_pid, q_camid, g_idx - int64_t[:] order = np.zeros(num_g, dtype=np.int64) - int64_t keep - - float[:] raw_cmc = np.zeros(num_g, dtype=np.float32) # binary vector, positions with value 1 are correct matches - float[:] masked_raw_cmc = np.zeros(num_g, dtype=np.float32) - float[:] cmc, masked_cmc - int64_t num_g_real, num_g_real_masked, rank_idx, rnd_idx - uint64_t meet_condition - float AP - int64_t[:] kept_g_pids, mask - - float num_rel - float[:] tmp_cmc = np.zeros(num_g, dtype=np.float32) - float tmp_cmc_sum - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - for g_idx in range(num_g): - order[g_idx] = indices[q_idx, g_idx] - num_g_real = 0 - meet_condition = 0 - kept_g_pids = np.zeros(num_g, dtype=np.int64) - - for g_idx in range(num_g): - if (g_pids[order[g_idx]] != q_pid) or (g_camids[order[g_idx]] != q_camid): - raw_cmc[num_g_real] = matches[q_idx][g_idx] - kept_g_pids[num_g_real] = g_pids[order[g_idx]] - num_g_real += 1 - if matches[q_idx][g_idx] > 1e-31: - meet_condition = 1 - - if not meet_condition: - # this condition is true when query identity does not appear in gallery - continue - - # cuhk03-specific setting - g_pids_dict = defaultdict(list) # overhead! - for g_idx in range(num_g_real): - g_pids_dict[kept_g_pids[g_idx]].append(g_idx) - - cmc = np.zeros(max_rank, dtype=np.float32) - for _ in range(num_repeats): - mask = np.zeros(num_g_real, dtype=np.int64) - - for _, idxs in g_pids_dict.items(): - # randomly sample one image for each gallery person - rnd_idx = np.random.choice(idxs) - #rnd_idx = idxs[0] # use deterministic for debugging - mask[rnd_idx] = 1 - - num_g_real_masked = 0 - for g_idx in range(num_g_real): - if mask[g_idx] == 1: - masked_raw_cmc[num_g_real_masked] = raw_cmc[g_idx] - num_g_real_masked += 1 - - masked_cmc = np.zeros(num_g, dtype=np.float32) - function_cumsum(masked_raw_cmc, masked_cmc, num_g_real_masked) - for g_idx in range(num_g_real_masked): - if masked_cmc[g_idx] > 1: - masked_cmc[g_idx] = 1 - - for rank_idx in range(max_rank): - cmc[rank_idx] += masked_cmc[rank_idx] / num_repeats - - for rank_idx in range(max_rank): - all_cmc[q_idx, rank_idx] = cmc[rank_idx] - # compute average precision - # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision - function_cumsum(raw_cmc, tmp_cmc, num_g_real) - num_rel = 0 - tmp_cmc_sum = 0 - for g_idx in range(num_g_real): - tmp_cmc_sum += (tmp_cmc[g_idx] / (g_idx + 1.)) * raw_cmc[g_idx] - num_rel += raw_cmc[g_idx] - all_AP[q_idx] = tmp_cmc_sum / num_rel - num_valid_q += 1. - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - # compute averaged cmc - cdef float[:] avg_cmc = np.zeros(max_rank, dtype=np.float32) - for rank_idx in range(max_rank): - for q_idx in range(num_q): - avg_cmc[rank_idx] += all_cmc[q_idx, rank_idx] - avg_cmc[rank_idx] /= num_valid_q - - cdef float mAP = 0 - for q_idx in range(num_q): - mAP += all_AP[q_idx] - mAP /= num_valid_q - - return np.asarray(avg_cmc).astype(np.float32), mAP - - -cpdef eval_market1501_cy(float[:,:] distmat, int64_t[:] q_pids, int64_t[:]g_pids, - int64_t[:]q_camids, int64_t[:]g_camids, int64_t max_rank): - - cdef int64_t num_q = distmat.shape[0] - cdef int64_t num_g = distmat.shape[1] - - if num_g < max_rank: - max_rank = num_g - print('Note: number of gallery samples is quite small, got {}'.format(num_g)) - - cdef: - int64_t[:,:] indices = np.argsort(distmat, axis=1) - int64_t[:,:] matches = (np.asarray(g_pids)[np.asarray(indices)] == np.asarray(q_pids)[:, np.newaxis]).astype(np.int64) - - float[:,:] all_cmc = np.zeros((num_q, max_rank), dtype=np.float32) - float[:] all_AP = np.zeros(num_q, dtype=np.float32) - float num_valid_q = 0. # number of valid query - - int64_t q_idx, q_pid, q_camid, g_idx - int64_t[:] order = np.zeros(num_g, dtype=np.int64) - int64_t keep - - float[:] raw_cmc = np.zeros(num_g, dtype=np.float32) # binary vector, positions with value 1 are correct matches - float[:] cmc = np.zeros(num_g, dtype=np.float32) - int64_t num_g_real, rank_idx - uint64_t meet_condition - - float num_rel - float[:] tmp_cmc = np.zeros(num_g, dtype=np.float32) - float tmp_cmc_sum - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - for g_idx in range(num_g): - order[g_idx] = indices[q_idx, g_idx] - num_g_real = 0 - meet_condition = 0 - - for g_idx in range(num_g): - if (g_pids[order[g_idx]] != q_pid) or (g_camids[order[g_idx]] != q_camid): - raw_cmc[num_g_real] = matches[q_idx][g_idx] - num_g_real += 1 - if matches[q_idx][g_idx] > 1e-31: - meet_condition = 1 - - if not meet_condition: - # this condition is true when query identity does not appear in gallery - continue - - # compute cmc - function_cumsum(raw_cmc, cmc, num_g_real) - for g_idx in range(num_g_real): - if cmc[g_idx] > 1: - cmc[g_idx] = 1 - - for rank_idx in range(max_rank): - all_cmc[q_idx, rank_idx] = cmc[rank_idx] - num_valid_q += 1. - - # compute average precision - # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision - function_cumsum(raw_cmc, tmp_cmc, num_g_real) - num_rel = 0 - tmp_cmc_sum = 0 - for g_idx in range(num_g_real): - tmp_cmc_sum += (tmp_cmc[g_idx] / (g_idx + 1.)) * raw_cmc[g_idx] - num_rel += raw_cmc[g_idx] - all_AP[q_idx] = tmp_cmc_sum / num_rel - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - # compute averaged cmc - cdef float[:] avg_cmc = np.zeros(max_rank, dtype=np.float32) - for rank_idx in range(max_rank): - for q_idx in range(num_q): - avg_cmc[rank_idx] += all_cmc[q_idx, rank_idx] - avg_cmc[rank_idx] /= num_valid_q - - cdef float mAP = 0 - for q_idx in range(num_q): - mAP += all_AP[q_idx] - mAP /= num_valid_q - - return np.asarray(avg_cmc).astype(np.float32), mAP - - -# Compute the cumulative sum -cdef void function_cumsum(cython.numeric[:] src, cython.numeric[:] dst, int64_t n): - cdef int64_t i - dst[0] = src[0] - for i in range(1, n): - dst[i] = src[i] + dst[i - 1] \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py deleted file mode 100644 index 2c2d996fcc..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py +++ /dev/null @@ -1,73 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -import numpy as np -from distutils.core import setup -from distutils.extension import Extension -from Cython.Build import cythonize - - -def numpy_include(): - try: - numpy_include = np.get_include() - except AttributeError: - numpy_include = np.get_numpy_include() - return numpy_include - - -ext_modules = [ - Extension( - 'rank_cy', - ['rank_cy.pyx'], - include_dirs=[numpy_include()], - ) -] - -setup( - name='Cython-based reid evaluation code', - ext_modules=cythonize(ext_modules) -) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py deleted file mode 100644 index 67e5f61114..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py +++ /dev/null @@ -1,130 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function -import sys -import numpy as np -import timeit -import os.path as osp - -from torchreid import metrics - -sys.path.insert(0, osp.dirname(osp.abspath(__file__)) + '/../../..') -""" -Test the speed of cython-based evaluation code. The speed improvements -can be much bigger when using the real reid data, which contains a larger -amount of query and gallery images. - -Note: you might encounter the following error: - 'AssertionError: Error: all query identities do not appear in gallery'. -This is normal because the inputs are random numbers. Just try again. -""" - -print('*** Compare running time ***') - -setup = ''' -import sys -import os.path as osp -import numpy as np -sys.path.insert(0, osp.dirname(osp.abspath(__file__)) + '/../../..') -from torchreid import metrics -num_q = 30 -num_g = 300 -max_rank = 5 -distmat = np.random.rand(num_q, num_g) * 20 -q_pids = np.random.randint(0, num_q, size=num_q) -g_pids = np.random.randint(0, num_g, size=num_g) -q_camids = np.random.randint(0, 5, size=num_q) -g_camids = np.random.randint(0, 5, size=num_g) -''' - -print('=> Using market1501\'s metric') -pytime = timeit.timeit( - 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=False)', - setup=setup, - number=20 -) -cytime = timeit.timeit( - 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=True)', - setup=setup, - number=20 -) -print('Python time: {} s'.format(pytime)) -print('Cython time: {} s'.format(cytime)) -print('Cython is {} times faster than python\n'.format(pytime / cytime)) - -print('=> Using cuhk03\'s metric') -pytime = timeit.timeit( - 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=True, use_cython=False)', - setup=setup, - number=20 -) -cytime = timeit.timeit( - 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=True, use_cython=True)', - setup=setup, - number=20 -) -print('Python time: {} s'.format(pytime)) -print('Cython time: {} s'.format(cytime)) -print('Cython is {} times faster than python\n'.format(pytime / cytime)) -""" -print("=> Check precision") - -num_q = 30 -num_g = 300 -max_rank = 5 -distmat = np.random.rand(num_q, num_g) * 20 -q_pids = np.random.randint(0, num_q, size=num_q) -g_pids = np.random.randint(0, num_g, size=num_g) -q_camids = np.random.randint(0, 5, size=num_q) -g_camids = np.random.randint(0, 5, size=num_g) - -cmc, mAP = evaluate(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=False) -print("Python:\nmAP = {} \ncmc = {}\n".format(mAP, cmc)) -cmc, mAP = evaluate(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=True) -print("Cython:\nmAP = {} \ncmc = {}\n".format(mAP, cmc)) -""" diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py deleted file mode 100644 index 7d0a09d988..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py +++ /dev/null @@ -1,166 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import -import torch - -from .pcb import * -from .mlfn import * -from .hacnn import * -from .osnet import * -from .senet import * -from .mudeep import * -from .nasnet import * -from .resnet import * -from .densenet import * -from .xception import * -from .osnet_ain import * -from .resnetmid import * -from .shufflenet import * -from .squeezenet import * -from .inceptionv4 import * -from .mobilenetv2 import * -from .resnet_ibn_a import * -from .resnet_ibn_b import * -from .shufflenetv2 import * -from .inceptionresnetv2 import * - -__model_factory = { - # image classification models - 'resnet18': resnet18, - 'resnet34': resnet34, - 'resnet50': resnet50, - 'resnet101': resnet101, - 'resnet152': resnet152, - 'resnext50_32x4d': resnext50_32x4d, - 'resnext101_32x8d': resnext101_32x8d, - 'resnet50_fc512': resnet50_fc512, - 'se_resnet50': se_resnet50, - 'se_resnet50_fc512': se_resnet50_fc512, - 'se_resnet101': se_resnet101, - 'se_resnext50_32x4d': se_resnext50_32x4d, - 'se_resnext101_32x4d': se_resnext101_32x4d, - 'densenet121': densenet121, - 'densenet169': densenet169, - 'densenet201': densenet201, - 'densenet161': densenet161, - 'densenet121_fc512': densenet121_fc512, - 'inceptionresnetv2': inceptionresnetv2, - 'inceptionv4': inceptionv4, - 'xception': xception, - 'resnet50_ibn_a': resnet50_ibn_a, - 'resnet50_ibn_b': resnet50_ibn_b, - # lightweight models - 'nasnsetmobile': nasnetamobile, - 'mobilenetv2_x1_0': mobilenetv2_x1_0, - 'mobilenetv2_x1_4': mobilenetv2_x1_4, - 'shufflenet': shufflenet, - 'squeezenet1_0': squeezenet1_0, - 'squeezenet1_0_fc512': squeezenet1_0_fc512, - 'squeezenet1_1': squeezenet1_1, - 'shufflenet_v2_x0_5': shufflenet_v2_x0_5, - 'shufflenet_v2_x1_0': shufflenet_v2_x1_0, - 'shufflenet_v2_x1_5': shufflenet_v2_x1_5, - 'shufflenet_v2_x2_0': shufflenet_v2_x2_0, - # reid-specific models - 'mudeep': MuDeep, - 'resnet50mid': resnet50mid, - 'hacnn': HACNN, - 'pcb_p6': pcb_p6, - 'pcb_p4': pcb_p4, - 'mlfn': mlfn, - 'osnet_x1_0': osnet_x1_0, - 'osnet_x0_75': osnet_x0_75, - 'osnet_x0_5': osnet_x0_5, - 'osnet_x0_25': osnet_x0_25, - 'osnet_ibn_x1_0': osnet_ibn_x1_0, - 'osnet_ain_x1_0': osnet_ain_x1_0 -} - - -def show_avai_models(): - """Displays available models. - - Examples:: - >>> from torchreid import models - >>> models.show_avai_models() - """ - print(list(__model_factory.keys())) - - -def build_model( - name, num_classes, loss='softmax', pretrained=True, use_gpu=True -): - """A function wrapper for building a model. - - Args: - name (str): model name. - num_classes (int): number of training identities. - loss (str, optional): loss function to optimize the model. Currently - supports "softmax" and "triplet". Default is "softmax". - pretrained (bool, optional): whether to load ImageNet-pretrained weights. - Default is True. - use_gpu (bool, optional): whether to use gpu. Default is True. - - Returns: - nn.Module - - Examples:: - >>> from torchreid import models - >>> model = models.build_model('resnet50', 751, loss='softmax') - """ - avai_models = list(__model_factory.keys()) - if name not in avai_models: - raise KeyError( - 'Unknown model: {}. Must be one of {}'.format(name, avai_models) - ) - return __model_factory[name]( - num_classes=num_classes, - loss=loss, - pretrained=pretrained, - use_gpu=use_gpu - ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py deleted file mode 100644 index 204aea72a1..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py +++ /dev/null @@ -1,425 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code source: https://github.com/pytorch/vision -""" -from __future__ import division, absolute_import -import re -from collections import OrderedDict -import torch -import torch.nn as nn -from torch.nn import functional as F -from torch.utils import model_zoo - -__all__ = [ - 'densenet121', 'densenet169', 'densenet201', 'densenet161', - 'densenet121_fc512' -] - -model_urls = { - 'densenet121': - 'https://download.pytorch.org/models/densenet121-a639ec97.pth', - 'densenet169': - 'https://download.pytorch.org/models/densenet169-b2777c0a.pth', - 'densenet201': - 'https://download.pytorch.org/models/densenet201-c1103571.pth', - 'densenet161': - 'https://download.pytorch.org/models/densenet161-8d451a50.pth', -} - - -class _DenseLayer(nn.Sequential): - - def __init__(self, num_input_features, growth_rate, bn_size, drop_rate): - super(_DenseLayer, self).__init__() - self.add_module('norm1', nn.BatchNorm2d(num_input_features)), - self.add_module('relu1', nn.ReLU(inplace=True)), - self.add_module( - 'conv1', - nn.Conv2d( - num_input_features, - bn_size * growth_rate, - kernel_size=1, - stride=1, - bias=False - ) - ), - self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate)), - self.add_module('relu2', nn.ReLU(inplace=True)), - self.add_module( - 'conv2', - nn.Conv2d( - bn_size * growth_rate, - growth_rate, - kernel_size=3, - stride=1, - padding=1, - bias=False - ) - ), - self.drop_rate = drop_rate - - def forward(self, x): - new_features = super(_DenseLayer, self).forward(x) - if self.drop_rate > 0: - new_features = F.dropout( - new_features, p=self.drop_rate, training=self.training - ) - return torch.cat([x, new_features], 1) - - -class _DenseBlock(nn.Sequential): - - def __init__( - self, num_layers, num_input_features, bn_size, growth_rate, drop_rate - ): - super(_DenseBlock, self).__init__() - for i in range(num_layers): - layer = _DenseLayer( - num_input_features + i*growth_rate, growth_rate, bn_size, - drop_rate - ) - self.add_module('denselayer%d' % (i+1), layer) - - -class _Transition(nn.Sequential): - - def __init__(self, num_input_features, num_output_features): - super(_Transition, self).__init__() - self.add_module('norm', nn.BatchNorm2d(num_input_features)) - self.add_module('relu', nn.ReLU(inplace=True)) - self.add_module( - 'conv', - nn.Conv2d( - num_input_features, - num_output_features, - kernel_size=1, - stride=1, - bias=False - ) - ) - self.add_module('pool', nn.AvgPool2d(kernel_size=2, stride=2)) - - -class DenseNet(nn.Module): - """Densely connected network. - - Reference: - Huang et al. Densely Connected Convolutional Networks. CVPR 2017. - - Public keys: - - ``densenet121``: DenseNet121. - - ``densenet169``: DenseNet169. - - ``densenet201``: DenseNet201. - - ``densenet161``: DenseNet161. - - ``densenet121_fc512``: DenseNet121 + FC. - """ - - def __init__( - self, - num_classes, - loss, - growth_rate=32, - block_config=(6, 12, 24, 16), - num_init_features=64, - bn_size=4, - drop_rate=0, - fc_dims=None, - dropout_p=None, - **kwargs - ): - - super(DenseNet, self).__init__() - self.loss = loss - - # First convolution - self.features = nn.Sequential( - OrderedDict( - [ - ( - 'conv0', - nn.Conv2d( - 3, - num_init_features, - kernel_size=7, - stride=2, - padding=3, - bias=False - ) - ), - ('norm0', nn.BatchNorm2d(num_init_features)), - ('relu0', nn.ReLU(inplace=True)), - ( - 'pool0', - nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - ), - ] - ) - ) - - # Each denseblock - num_features = num_init_features - for i, num_layers in enumerate(block_config): - block = _DenseBlock( - num_layers=num_layers, - num_input_features=num_features, - bn_size=bn_size, - growth_rate=growth_rate, - drop_rate=drop_rate - ) - self.features.add_module('denseblock%d' % (i+1), block) - num_features = num_features + num_layers*growth_rate - if i != len(block_config) - 1: - trans = _Transition( - num_input_features=num_features, - num_output_features=num_features // 2 - ) - self.features.add_module('transition%d' % (i+1), trans) - num_features = num_features // 2 - - # Final batch norm - self.features.add_module('norm5', nn.BatchNorm2d(num_features)) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.feature_dim = num_features - self.fc = self._construct_fc_layer(fc_dims, num_features, dropout_p) - - # Linear layer - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer. - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x): - f = self.features(x) - f = F.relu(f, inplace=True) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - - # '.'s are no longer allowed in module names, but pervious _DenseLayer - # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. - # They are also in the checkpoints in model_urls. This pattern is used - # to find such keys. - pattern = re.compile( - r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$' - ) - for key in list(pretrain_dict.keys()): - res = pattern.match(key) - if res: - new_key = res.group(1) + res.group(2) - pretrain_dict[new_key] = pretrain_dict[key] - del pretrain_dict[key] - - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -""" -Dense network configurations: --- -densenet121: num_init_features=64, growth_rate=32, block_config=(6, 12, 24, 16) -densenet169: num_init_features=64, growth_rate=32, block_config=(6, 12, 32, 32) -densenet201: num_init_features=64, growth_rate=32, block_config=(6, 12, 48, 32) -densenet161: num_init_features=96, growth_rate=48, block_config=(6, 12, 36, 24) -""" - - -def densenet121(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 24, 16), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet121']) - return model - - -def densenet169(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 32, 32), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet169']) - return model - - -def densenet201(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 48, 32), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet201']) - return model - - -def densenet161(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=96, - growth_rate=48, - block_config=(6, 12, 36, 24), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet161']) - return model - - -def densenet121_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 24, 16), - fc_dims=[512], - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet121']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py deleted file mode 100644 index 27dae2fa28..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py +++ /dev/null @@ -1,461 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -from torch import nn -from torch.nn import functional as F - -__all__ = ['HACNN'] - - -class ConvBlock(nn.Module): - """Basic convolutional block. - - convolution + batch normalization + relu. - - Args: - in_c (int): number of input channels. - out_c (int): number of output channels. - k (int or tuple): kernel size. - s (int or tuple): stride. - p (int or tuple): padding. - """ - - def __init__(self, in_c, out_c, k, s=1, p=0): - super(ConvBlock, self).__init__() - self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) - self.bn = nn.BatchNorm2d(out_c) - - def forward(self, x): - return F.relu(self.bn(self.conv(x))) - - -class InceptionA(nn.Module): - - def __init__(self, in_channels, out_channels): - super(InceptionA, self).__init__() - mid_channels = out_channels // 4 - - self.stream1 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream2 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream3 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream4 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1), - ConvBlock(in_channels, mid_channels, 1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - s4 = self.stream4(x) - y = torch.cat([s1, s2, s3, s4], dim=1) - return y - - -class InceptionB(nn.Module): - - def __init__(self, in_channels, out_channels): - super(InceptionB, self).__init__() - mid_channels = out_channels // 4 - - self.stream1 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), - ) - self.stream2 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), - ) - self.stream3 = nn.Sequential( - nn.MaxPool2d(3, stride=2, padding=1), - ConvBlock(in_channels, mid_channels * 2, 1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - y = torch.cat([s1, s2, s3], dim=1) - return y - - -class SpatialAttn(nn.Module): - """Spatial Attention (Sec. 3.1.I.1)""" - - def __init__(self): - super(SpatialAttn, self).__init__() - self.conv1 = ConvBlock(1, 1, 3, s=2, p=1) - self.conv2 = ConvBlock(1, 1, 1) - - def forward(self, x): - # global cross-channel averaging - x = x.mean(1, keepdim=True) - # 3-by-3 conv - x = self.conv1(x) - # bilinear resizing - x = F.upsample( - x, (x.size(2) * 2, x.size(3) * 2), - mode='bilinear', - align_corners=True - ) - # scaling conv - x = self.conv2(x) - return x - - -class ChannelAttn(nn.Module): - """Channel Attention (Sec. 3.1.I.2)""" - - def __init__(self, in_channels, reduction_rate=16): - super(ChannelAttn, self).__init__() - assert in_channels % reduction_rate == 0 - self.conv1 = ConvBlock(in_channels, in_channels // reduction_rate, 1) - self.conv2 = ConvBlock(in_channels // reduction_rate, in_channels, 1) - - def forward(self, x): - # squeeze operation (global average pooling) - x = F.avg_pool2d(x, x.size()[2:]) - # excitation operation (2 conv layers) - x = self.conv1(x) - x = self.conv2(x) - return x - - -class SoftAttn(nn.Module): - """Soft Attention (Sec. 3.1.I) - - Aim: Spatial Attention + Channel Attention - - Output: attention maps with shape identical to input. - """ - - def __init__(self, in_channels): - super(SoftAttn, self).__init__() - self.spatial_attn = SpatialAttn() - self.channel_attn = ChannelAttn(in_channels) - self.conv = ConvBlock(in_channels, in_channels, 1) - - def forward(self, x): - y_spatial = self.spatial_attn(x) - y_channel = self.channel_attn(x) - y = y_spatial * y_channel - y = torch.sigmoid(self.conv(y)) - return y - - -class HardAttn(nn.Module): - """Hard Attention (Sec. 3.1.II)""" - - def __init__(self, in_channels): - super(HardAttn, self).__init__() - self.fc = nn.Linear(in_channels, 4 * 2) - self.init_params() - - def init_params(self): - self.fc.weight.data.zero_() - self.fc.bias.data.copy_( - torch.tensor( - [0, -0.75, 0, -0.25, 0, 0.25, 0, 0.75], dtype=torch.float - ) - ) - - def forward(self, x): - # squeeze operation (global average pooling) - x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), x.size(1)) - # predict transformation parameters - theta = torch.tanh(self.fc(x)) - theta = theta.view(-1, 4, 2) - return theta - - -class HarmAttn(nn.Module): - """Harmonious Attention (Sec. 3.1)""" - - def __init__(self, in_channels): - super(HarmAttn, self).__init__() - self.soft_attn = SoftAttn(in_channels) - self.hard_attn = HardAttn(in_channels) - - def forward(self, x): - y_soft_attn = self.soft_attn(x) - theta = self.hard_attn(x) - return y_soft_attn, theta - - -class HACNN(nn.Module): - """Harmonious Attention Convolutional Neural Network. - - Reference: - Li et al. Harmonious Attention Network for Person Re-identification. CVPR 2018. - - Public keys: - - ``hacnn``: HACNN. - """ - - # Args: - # num_classes (int): number of classes to predict - # nchannels (list): number of channels AFTER concatenation - # feat_dim (int): feature dimension for a single stream - # learn_region (bool): whether to learn region features (i.e. local branch) - - def __init__( - self, - num_classes, - loss='softmax', - nchannels=[128, 256, 384], - feat_dim=512, - learn_region=True, - use_gpu=True, - **kwargs - ): - super(HACNN, self).__init__() - self.loss = loss - self.learn_region = learn_region - self.use_gpu = use_gpu - - self.conv = ConvBlock(3, 32, 3, s=2, p=1) - - # Construct Inception + HarmAttn blocks - # ============== Block 1 ============== - self.inception1 = nn.Sequential( - InceptionA(32, nchannels[0]), - InceptionB(nchannels[0], nchannels[0]), - ) - self.ha1 = HarmAttn(nchannels[0]) - - # ============== Block 2 ============== - self.inception2 = nn.Sequential( - InceptionA(nchannels[0], nchannels[1]), - InceptionB(nchannels[1], nchannels[1]), - ) - self.ha2 = HarmAttn(nchannels[1]) - - # ============== Block 3 ============== - self.inception3 = nn.Sequential( - InceptionA(nchannels[1], nchannels[2]), - InceptionB(nchannels[2], nchannels[2]), - ) - self.ha3 = HarmAttn(nchannels[2]) - - self.fc_global = nn.Sequential( - nn.Linear(nchannels[2], feat_dim), - nn.BatchNorm1d(feat_dim), - nn.ReLU(), - ) - self.classifier_global = nn.Linear(feat_dim, num_classes) - - if self.learn_region: - self.init_scale_factors() - self.local_conv1 = InceptionB(32, nchannels[0]) - self.local_conv2 = InceptionB(nchannels[0], nchannels[1]) - self.local_conv3 = InceptionB(nchannels[1], nchannels[2]) - self.fc_local = nn.Sequential( - nn.Linear(nchannels[2] * 4, feat_dim), - nn.BatchNorm1d(feat_dim), - nn.ReLU(), - ) - self.classifier_local = nn.Linear(feat_dim, num_classes) - self.feat_dim = feat_dim * 2 - else: - self.feat_dim = feat_dim - - def init_scale_factors(self): - # initialize scale factors (s_w, s_h) for four regions - self.scale_factors = [] - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - - def stn(self, x, theta): - """Performs spatial transform - - x: (batch, channel, height, width) - theta: (batch, 2, 3) - """ - grid = F.affine_grid(theta, x.size()) - x = F.grid_sample(x, grid) - return x - - def transform_theta(self, theta_i, region_idx): - """Transforms theta to include (s_w, s_h), resulting in (batch, 2, 3)""" - scale_factors = self.scale_factors[region_idx] - theta = torch.zeros(theta_i.size(0), 2, 3) - theta[:, :, :2] = scale_factors - theta[:, :, -1] = theta_i - if self.use_gpu: - theta = theta.cuda() - return theta - - def forward(self, x): - assert x.size(2) == 160 and x.size(3) == 64, \ - 'Input size does not match, expected (160, 64) but got ({}, {})'.format(x.size(2), x.size(3)) - x = self.conv(x) - - # ============== Block 1 ============== - # global branch - x1 = self.inception1(x) - x1_attn, x1_theta = self.ha1(x1) - x1_out = x1 * x1_attn - # local branch - if self.learn_region: - x1_local_list = [] - for region_idx in range(4): - x1_theta_i = x1_theta[:, region_idx, :] - x1_theta_i = self.transform_theta(x1_theta_i, region_idx) - x1_trans_i = self.stn(x, x1_theta_i) - x1_trans_i = F.upsample( - x1_trans_i, (24, 28), mode='bilinear', align_corners=True - ) - x1_local_i = self.local_conv1(x1_trans_i) - x1_local_list.append(x1_local_i) - - # ============== Block 2 ============== - # Block 2 - # global branch - x2 = self.inception2(x1_out) - x2_attn, x2_theta = self.ha2(x2) - x2_out = x2 * x2_attn - # local branch - if self.learn_region: - x2_local_list = [] - for region_idx in range(4): - x2_theta_i = x2_theta[:, region_idx, :] - x2_theta_i = self.transform_theta(x2_theta_i, region_idx) - x2_trans_i = self.stn(x1_out, x2_theta_i) - x2_trans_i = F.upsample( - x2_trans_i, (12, 14), mode='bilinear', align_corners=True - ) - x2_local_i = x2_trans_i + x1_local_list[region_idx] - x2_local_i = self.local_conv2(x2_local_i) - x2_local_list.append(x2_local_i) - - # ============== Block 3 ============== - # Block 3 - # global branch - x3 = self.inception3(x2_out) - x3_attn, x3_theta = self.ha3(x3) - x3_out = x3 * x3_attn - # local branch - if self.learn_region: - x3_local_list = [] - for region_idx in range(4): - x3_theta_i = x3_theta[:, region_idx, :] - x3_theta_i = self.transform_theta(x3_theta_i, region_idx) - x3_trans_i = self.stn(x2_out, x3_theta_i) - x3_trans_i = F.upsample( - x3_trans_i, (6, 7), mode='bilinear', align_corners=True - ) - x3_local_i = x3_trans_i + x2_local_list[region_idx] - x3_local_i = self.local_conv3(x3_local_i) - x3_local_list.append(x3_local_i) - - # ============== Feature generation ============== - # global branch - x_global = F.avg_pool2d(x3_out, - x3_out.size()[2:] - ).view(x3_out.size(0), x3_out.size(1)) - x_global = self.fc_global(x_global) - # local branch - if self.learn_region: - x_local_list = [] - for region_idx in range(4): - x_local_i = x3_local_list[region_idx] - x_local_i = F.avg_pool2d(x_local_i, - x_local_i.size()[2:] - ).view(x_local_i.size(0), -1) - x_local_list.append(x_local_i) - x_local = torch.cat(x_local_list, 1) - x_local = self.fc_local(x_local) - - if not self.training: - # l2 normalization before concatenation - if self.learn_region: - x_global = x_global / x_global.norm(p=2, dim=1, keepdim=True) - x_local = x_local / x_local.norm(p=2, dim=1, keepdim=True) - return torch.cat([x_global, x_local], 1) - else: - return x_global - - prelogits_global = self.classifier_global(x_global) - if self.learn_region: - prelogits_local = self.classifier_local(x_local) - - if self.loss == 'softmax': - if self.learn_region: - return (prelogits_global, prelogits_local) - else: - return prelogits_global - - elif self.loss == 'triplet': - if self.learn_region: - return (prelogits_global, prelogits_local), (x_global, x_local) - else: - return prelogits_global, x_global - - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py deleted file mode 100644 index f9d62718b3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py +++ /dev/null @@ -1,406 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['inceptionresnetv2'] - -pretrained_settings = { - 'inceptionresnetv2': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000 - }, - 'imagenet+background': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1001 - } - } -} - - -class BasicConv2d(nn.Module): - - def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): - super(BasicConv2d, self).__init__() - self.conv = nn.Conv2d( - in_planes, - out_planes, - kernel_size=kernel_size, - stride=stride, - padding=padding, - bias=False - ) # verify bias false - self.bn = nn.BatchNorm2d( - out_planes, - eps=0.001, # value found in tensorflow - momentum=0.1, # default pytorch value - affine=True - ) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Mixed_5b(nn.Module): - - def __init__(self): - super(Mixed_5b, self).__init__() - - self.branch0 = BasicConv2d(192, 96, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(192, 48, kernel_size=1, stride=1), - BasicConv2d(48, 64, kernel_size=5, stride=1, padding=2) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(192, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), - BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(192, 64, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Block35(nn.Module): - - def __init__(self, scale=1.0): - super(Block35, self).__init__() - - self.scale = scale - - self.branch0 = BasicConv2d(320, 32, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(320, 32, kernel_size=1, stride=1), - BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(320, 32, kernel_size=1, stride=1), - BasicConv2d(32, 48, kernel_size=3, stride=1, padding=1), - BasicConv2d(48, 64, kernel_size=3, stride=1, padding=1) - ) - - self.conv2d = nn.Conv2d(128, 320, kernel_size=1, stride=1) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - out = self.conv2d(out) - out = out * self.scale + x - out = self.relu(out) - return out - - -class Mixed_6a(nn.Module): - - def __init__(self): - super(Mixed_6a, self).__init__() - - self.branch0 = BasicConv2d(320, 384, kernel_size=3, stride=2) - - self.branch1 = nn.Sequential( - BasicConv2d(320, 256, kernel_size=1, stride=1), - BasicConv2d(256, 256, kernel_size=3, stride=1, padding=1), - BasicConv2d(256, 384, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Block17(nn.Module): - - def __init__(self, scale=1.0): - super(Block17, self).__init__() - - self.scale = scale - - self.branch0 = BasicConv2d(1088, 192, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(1088, 128, kernel_size=1, stride=1), - BasicConv2d( - 128, 160, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 160, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) - ) - ) - - self.conv2d = nn.Conv2d(384, 1088, kernel_size=1, stride=1) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - out = self.conv2d(out) - out = out * self.scale + x - out = self.relu(out) - return out - - -class Mixed_7a(nn.Module): - - def __init__(self): - super(Mixed_7a, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 384, kernel_size=3, stride=2) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 288, kernel_size=3, stride=2) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 288, kernel_size=3, stride=1, padding=1), - BasicConv2d(288, 320, kernel_size=3, stride=2) - ) - - self.branch3 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Block8(nn.Module): - - def __init__(self, scale=1.0, noReLU=False): - super(Block8, self).__init__() - - self.scale = scale - self.noReLU = noReLU - - self.branch0 = BasicConv2d(2080, 192, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(2080, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 224, kernel_size=(1, 3), stride=1, padding=(0, 1) - ), - BasicConv2d( - 224, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - ) - - self.conv2d = nn.Conv2d(448, 2080, kernel_size=1, stride=1) - if not self.noReLU: - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - out = self.conv2d(out) - out = out * self.scale + x - if not self.noReLU: - out = self.relu(out) - return out - - -# ---------------- -# Model Definition -# ---------------- -class InceptionResNetV2(nn.Module): - """Inception-ResNet-V2. - - Reference: - Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual - Connections on Learning. AAAI 2017. - - Public keys: - - ``inceptionresnetv2``: Inception-ResNet-V2. - """ - - def __init__(self, num_classes, loss='softmax', **kwargs): - super(InceptionResNetV2, self).__init__() - self.loss = loss - - # Modules - self.conv2d_1a = BasicConv2d(3, 32, kernel_size=3, stride=2) - self.conv2d_2a = BasicConv2d(32, 32, kernel_size=3, stride=1) - self.conv2d_2b = BasicConv2d( - 32, 64, kernel_size=3, stride=1, padding=1 - ) - self.maxpool_3a = nn.MaxPool2d(3, stride=2) - self.conv2d_3b = BasicConv2d(64, 80, kernel_size=1, stride=1) - self.conv2d_4a = BasicConv2d(80, 192, kernel_size=3, stride=1) - self.maxpool_5a = nn.MaxPool2d(3, stride=2) - self.mixed_5b = Mixed_5b() - self.repeat = nn.Sequential( - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17) - ) - self.mixed_6a = Mixed_6a() - self.repeat_1 = nn.Sequential( - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10) - ) - self.mixed_7a = Mixed_7a() - self.repeat_2 = nn.Sequential( - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20) - ) - - self.block8 = Block8(noReLU=True) - self.conv2d_7b = BasicConv2d(2080, 1536, kernel_size=1, stride=1) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.classifier = nn.Linear(1536, num_classes) - - def load_imagenet_weights(self): - settings = pretrained_settings['inceptionresnetv2']['imagenet'] - pretrain_dict = model_zoo.load_url(settings['url']) - model_dict = self.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - self.load_state_dict(model_dict) - - def featuremaps(self, x): - x = self.conv2d_1a(x) - x = self.conv2d_2a(x) - x = self.conv2d_2b(x) - x = self.maxpool_3a(x) - x = self.conv2d_3b(x) - x = self.conv2d_4a(x) - x = self.maxpool_5a(x) - x = self.mixed_5b(x) - x = self.repeat(x) - x = self.mixed_6a(x) - x = self.repeat_1(x) - x = self.mixed_7a(x) - x = self.repeat_2(x) - x = self.block8(x) - x = self.conv2d_7b(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def inceptionresnetv2(num_classes, loss='softmax', pretrained=True, **kwargs): - model = InceptionResNetV2(num_classes=num_classes, loss=loss, **kwargs) - if pretrained: - model.load_imagenet_weights() - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py deleted file mode 100644 index 32a847a88f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py +++ /dev/null @@ -1,428 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['inceptionv4'] -""" -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" - -pretrained_settings = { - 'inceptionv4': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000 - }, - 'imagenet+background': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1001 - } - } -} - - -class BasicConv2d(nn.Module): - - def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): - super(BasicConv2d, self).__init__() - self.conv = nn.Conv2d( - in_planes, - out_planes, - kernel_size=kernel_size, - stride=stride, - padding=padding, - bias=False - ) # verify bias false - self.bn = nn.BatchNorm2d( - out_planes, - eps=0.001, # value found in tensorflow - momentum=0.1, # default pytorch value - affine=True - ) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Mixed_3a(nn.Module): - - def __init__(self): - super(Mixed_3a, self).__init__() - self.maxpool = nn.MaxPool2d(3, stride=2) - self.conv = BasicConv2d(64, 96, kernel_size=3, stride=2) - - def forward(self, x): - x0 = self.maxpool(x) - x1 = self.conv(x) - out = torch.cat((x0, x1), 1) - return out - - -class Mixed_4a(nn.Module): - - def __init__(self): - super(Mixed_4a, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(160, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(160, 64, kernel_size=1, stride=1), - BasicConv2d(64, 64, kernel_size=(1, 7), stride=1, padding=(0, 3)), - BasicConv2d(64, 64, kernel_size=(7, 1), stride=1, padding=(3, 0)), - BasicConv2d(64, 96, kernel_size=(3, 3), stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - return out - - -class Mixed_5a(nn.Module): - - def __init__(self): - super(Mixed_5a, self).__init__() - self.conv = BasicConv2d(192, 192, kernel_size=3, stride=2) - self.maxpool = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.conv(x) - x1 = self.maxpool(x) - out = torch.cat((x0, x1), 1) - return out - - -class Inception_A(nn.Module): - - def __init__(self): - super(Inception_A, self).__init__() - self.branch0 = BasicConv2d(384, 96, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(384, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(384, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), - BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(384, 96, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Reduction_A(nn.Module): - - def __init__(self): - super(Reduction_A, self).__init__() - self.branch0 = BasicConv2d(384, 384, kernel_size=3, stride=2) - - self.branch1 = nn.Sequential( - BasicConv2d(384, 192, kernel_size=1, stride=1), - BasicConv2d(192, 224, kernel_size=3, stride=1, padding=1), - BasicConv2d(224, 256, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Inception_B(nn.Module): - - def __init__(self): - super(Inception_B, self).__init__() - self.branch0 = BasicConv2d(1024, 384, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 224, 256, kernel_size=(7, 1), stride=1, padding=(3, 0) - ) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), - BasicConv2d( - 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 224, 224, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), - BasicConv2d( - 224, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) - ) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(1024, 128, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Reduction_B(nn.Module): - - def __init__(self): - super(Reduction_B, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d(192, 192, kernel_size=3, stride=2) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(1024, 256, kernel_size=1, stride=1), - BasicConv2d( - 256, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 256, 320, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), BasicConv2d(320, 320, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Inception_C(nn.Module): - - def __init__(self): - super(Inception_C, self).__init__() - - self.branch0 = BasicConv2d(1536, 256, kernel_size=1, stride=1) - - self.branch1_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) - self.branch1_1a = BasicConv2d( - 384, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch1_1b = BasicConv2d( - 384, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - - self.branch2_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) - self.branch2_1 = BasicConv2d( - 384, 448, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - self.branch2_2 = BasicConv2d( - 448, 512, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch2_3a = BasicConv2d( - 512, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch2_3b = BasicConv2d( - 512, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(1536, 256, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - - x1_0 = self.branch1_0(x) - x1_1a = self.branch1_1a(x1_0) - x1_1b = self.branch1_1b(x1_0) - x1 = torch.cat((x1_1a, x1_1b), 1) - - x2_0 = self.branch2_0(x) - x2_1 = self.branch2_1(x2_0) - x2_2 = self.branch2_2(x2_1) - x2_3a = self.branch2_3a(x2_2) - x2_3b = self.branch2_3b(x2_2) - x2 = torch.cat((x2_3a, x2_3b), 1) - - x3 = self.branch3(x) - - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class InceptionV4(nn.Module): - """Inception-v4. - - Reference: - Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual - Connections on Learning. AAAI 2017. - - Public keys: - - ``inceptionv4``: InceptionV4. - """ - - def __init__(self, num_classes, loss, **kwargs): - super(InceptionV4, self).__init__() - self.loss = loss - - self.features = nn.Sequential( - BasicConv2d(3, 32, kernel_size=3, stride=2), - BasicConv2d(32, 32, kernel_size=3, stride=1), - BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1), - Mixed_3a(), - Mixed_4a(), - Mixed_5a(), - Inception_A(), - Inception_A(), - Inception_A(), - Inception_A(), - Reduction_A(), # Mixed_6a - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Reduction_B(), # Mixed_7a - Inception_C(), - Inception_C(), - Inception_C() - ) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.classifier = nn.Linear(1536, num_classes) - - def forward(self, x): - f = self.features(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def inceptionv4(num_classes, loss='softmax', pretrained=True, **kwargs): - model = InceptionV4(num_classes, loss, **kwargs) - if pretrained: - model_url = pretrained_settings['inceptionv4']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py deleted file mode 100644 index 0e538241f7..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py +++ /dev/null @@ -1,316 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['mlfn'] - -model_urls = { - # training epoch = 5, top1 = 51.6 - 'imagenet': - 'https://mega.nz/#!YHxAhaxC!yu9E6zWl0x5zscSouTdbZu8gdFFytDdl-RAdD2DEfpk', -} - - -class MLFNBlock(nn.Module): - - def __init__( - self, in_channels, out_channels, stride, fsm_channels, groups=32 - ): - super(MLFNBlock, self).__init__() - self.groups = groups - mid_channels = out_channels // 2 - - # Factor Modules - self.fm_conv1 = nn.Conv2d(in_channels, mid_channels, 1, bias=False) - self.fm_bn1 = nn.BatchNorm2d(mid_channels) - self.fm_conv2 = nn.Conv2d( - mid_channels, - mid_channels, - 3, - stride=stride, - padding=1, - bias=False, - groups=self.groups - ) - self.fm_bn2 = nn.BatchNorm2d(mid_channels) - self.fm_conv3 = nn.Conv2d(mid_channels, out_channels, 1, bias=False) - self.fm_bn3 = nn.BatchNorm2d(out_channels) - - # Factor Selection Module - self.fsm = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - nn.Conv2d(in_channels, fsm_channels[0], 1), - nn.BatchNorm2d(fsm_channels[0]), - nn.ReLU(inplace=True), - nn.Conv2d(fsm_channels[0], fsm_channels[1], 1), - nn.BatchNorm2d(fsm_channels[1]), - nn.ReLU(inplace=True), - nn.Conv2d(fsm_channels[1], self.groups, 1), - nn.BatchNorm2d(self.groups), - nn.Sigmoid(), - ) - - self.downsample = None - if in_channels != out_channels or stride > 1: - self.downsample = nn.Sequential( - nn.Conv2d( - in_channels, out_channels, 1, stride=stride, bias=False - ), - nn.BatchNorm2d(out_channels), - ) - - def forward(self, x): - residual = x - s = self.fsm(x) - - # reduce dimension - x = self.fm_conv1(x) - x = self.fm_bn1(x) - x = F.relu(x, inplace=True) - - # group convolution - x = self.fm_conv2(x) - x = self.fm_bn2(x) - x = F.relu(x, inplace=True) - - # factor selection - b, c = x.size(0), x.size(1) - n = c // self.groups - ss = s.repeat(1, n, 1, 1) # from (b, g, 1, 1) to (b, g*n=c, 1, 1) - ss = ss.view(b, n, self.groups, 1, 1) - ss = ss.permute(0, 2, 1, 3, 4).contiguous() - ss = ss.view(b, c, 1, 1) - x = ss * x - - # recover dimension - x = self.fm_conv3(x) - x = self.fm_bn3(x) - x = F.relu(x, inplace=True) - - if self.downsample is not None: - residual = self.downsample(residual) - - return F.relu(residual + x, inplace=True), s - - -class MLFN(nn.Module): - """Multi-Level Factorisation Net. - - Reference: - Chang et al. Multi-Level Factorisation Net for - Person Re-Identification. CVPR 2018. - - Public keys: - - ``mlfn``: MLFN (Multi-Level Factorisation Net). - """ - - def __init__( - self, - num_classes, - loss='softmax', - groups=32, - channels=[64, 256, 512, 1024, 2048], - embed_dim=1024, - **kwargs - ): - super(MLFN, self).__init__() - self.loss = loss - self.groups = groups - - # first convolutional layer - self.conv1 = nn.Conv2d(3, channels[0], 7, stride=2, padding=3) - self.bn1 = nn.BatchNorm2d(channels[0]) - self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) - - # main body - self.feature = nn.ModuleList( - [ - # layer 1-3 - MLFNBlock(channels[0], channels[1], 1, [128, 64], self.groups), - MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), - MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), - # layer 4-7 - MLFNBlock( - channels[1], channels[2], 2, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - # layer 8-13 - MLFNBlock( - channels[2], channels[3], 2, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - # layer 14-16 - MLFNBlock( - channels[3], channels[4], 2, [512, 128], self.groups - ), - MLFNBlock( - channels[4], channels[4], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[4], channels[4], 1, [512, 128], self.groups - ), - ] - ) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - - # projection functions - self.fc_x = nn.Sequential( - nn.Conv2d(channels[4], embed_dim, 1, bias=False), - nn.BatchNorm2d(embed_dim), - nn.ReLU(inplace=True), - ) - self.fc_s = nn.Sequential( - nn.Conv2d(self.groups * 16, embed_dim, 1, bias=False), - nn.BatchNorm2d(embed_dim), - nn.ReLU(inplace=True), - ) - - self.classifier = nn.Linear(embed_dim, num_classes) - - self.init_params() - - def init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = F.relu(x, inplace=True) - x = self.maxpool(x) - - s_hat = [] - for block in self.feature: - x, s = block(x) - s_hat.append(s) - s_hat = torch.cat(s_hat, 1) - - x = self.global_avgpool(x) - x = self.fc_x(x) - s_hat = self.fc_s(s_hat) - - v = (x+s_hat) * 0.5 - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def mlfn(num_classes, loss='softmax', pretrained=True, **kwargs): - model = MLFN(num_classes, loss, **kwargs) - if pretrained: - # init_pretrained_weights(model, model_urls['imagenet']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['imagenet']) - ) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py deleted file mode 100644 index 690dade1bb..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py +++ /dev/null @@ -1,321 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['mobilenetv2_x1_0', 'mobilenetv2_x1_4'] - -model_urls = { - # 1.0: top-1 71.3 - 'mobilenetv2_x1_0': - 'https://mega.nz/#!NKp2wAIA!1NH1pbNzY_M2hVk_hdsxNM1NUOWvvGPHhaNr-fASF6c', - # 1.4: top-1 73.9 - 'mobilenetv2_x1_4': - 'https://mega.nz/#!RGhgEIwS!xN2s2ZdyqI6vQ3EwgmRXLEW3khr9tpXg96G9SUJugGk', -} - - -class ConvBlock(nn.Module): - """Basic convolutional block. - - convolution (bias discarded) + batch normalization + relu6. - - Args: - in_c (int): number of input channels. - out_c (int): number of output channels. - k (int or tuple): kernel size. - s (int or tuple): stride. - p (int or tuple): padding. - g (int): number of blocked connections from input channels - to output channels (default: 1). - """ - - def __init__(self, in_c, out_c, k, s=1, p=0, g=1): - super(ConvBlock, self).__init__() - self.conv = nn.Conv2d( - in_c, out_c, k, stride=s, padding=p, bias=False, groups=g - ) - self.bn = nn.BatchNorm2d(out_c) - - def forward(self, x): - return F.relu6(self.bn(self.conv(x))) - - -class Bottleneck(nn.Module): - - def __init__(self, in_channels, out_channels, expansion_factor, stride=1): - super(Bottleneck, self).__init__() - mid_channels = in_channels * expansion_factor - self.use_residual = stride == 1 and in_channels == out_channels - self.conv1 = ConvBlock(in_channels, mid_channels, 1) - self.dwconv2 = ConvBlock( - mid_channels, mid_channels, 3, stride, 1, g=mid_channels - ) - self.conv3 = nn.Sequential( - nn.Conv2d(mid_channels, out_channels, 1, bias=False), - nn.BatchNorm2d(out_channels), - ) - - def forward(self, x): - m = self.conv1(x) - m = self.dwconv2(m) - m = self.conv3(m) - if self.use_residual: - return x + m - else: - return m - - -class MobileNetV2(nn.Module): - """MobileNetV2. - - Reference: - Sandler et al. MobileNetV2: Inverted Residuals and - Linear Bottlenecks. CVPR 2018. - - Public keys: - - ``mobilenetv2_x1_0``: MobileNetV2 x1.0. - - ``mobilenetv2_x1_4``: MobileNetV2 x1.4. - """ - - def __init__( - self, - num_classes, - width_mult=1, - loss='softmax', - fc_dims=None, - dropout_p=None, - **kwargs - ): - super(MobileNetV2, self).__init__() - self.loss = loss - self.in_channels = int(32 * width_mult) - self.feature_dim = int(1280 * width_mult) if width_mult > 1 else 1280 - - # construct layers - self.conv1 = ConvBlock(3, self.in_channels, 3, s=2, p=1) - self.conv2 = self._make_layer( - Bottleneck, 1, int(16 * width_mult), 1, 1 - ) - self.conv3 = self._make_layer( - Bottleneck, 6, int(24 * width_mult), 2, 2 - ) - self.conv4 = self._make_layer( - Bottleneck, 6, int(32 * width_mult), 3, 2 - ) - self.conv5 = self._make_layer( - Bottleneck, 6, int(64 * width_mult), 4, 2 - ) - self.conv6 = self._make_layer( - Bottleneck, 6, int(96 * width_mult), 3, 1 - ) - self.conv7 = self._make_layer( - Bottleneck, 6, int(160 * width_mult), 3, 2 - ) - self.conv8 = self._make_layer( - Bottleneck, 6, int(320 * width_mult), 1, 1 - ) - self.conv9 = ConvBlock(self.in_channels, self.feature_dim, 1) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = self._construct_fc_layer( - fc_dims, self.feature_dim, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _make_layer(self, block, t, c, n, s): - # t: expansion factor - # c: output channels - # n: number of blocks - # s: stride for first layer - layers = [] - layers.append(block(self.in_channels, c, t, s)) - self.in_channels = c - for i in range(1, n): - layers.append(block(self.in_channels, c, t)) - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer. - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.conv2(x) - x = self.conv3(x) - x = self.conv4(x) - x = self.conv5(x) - x = self.conv6(x) - x = self.conv7(x) - x = self.conv8(x) - x = self.conv9(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def mobilenetv2_x1_0(num_classes, loss, pretrained=True, **kwargs): - model = MobileNetV2( - num_classes, - loss=loss, - width_mult=1, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - # init_pretrained_weights(model, model_urls['mobilenetv2_x1_0']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['mobilenetv2_x1_0']) - ) - return model - - -def mobilenetv2_x1_4(num_classes, loss, pretrained=True, **kwargs): - model = MobileNetV2( - num_classes, - loss=loss, - width_mult=1.4, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - # init_pretrained_weights(model, model_urls['mobilenetv2_x1_4']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['mobilenetv2_x1_4']) - ) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py deleted file mode 100644 index a2ec649d36..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py +++ /dev/null @@ -1,253 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -from torch import nn -from torch.nn import functional as F - -__all__ = ['MuDeep'] - - -class ConvBlock(nn.Module): - """Basic convolutional block. - - convolution + batch normalization + relu. - - Args: - in_c (int): number of input channels. - out_c (int): number of output channels. - k (int or tuple): kernel size. - s (int or tuple): stride. - p (int or tuple): padding. - """ - - def __init__(self, in_c, out_c, k, s, p): - super(ConvBlock, self).__init__() - self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) - self.bn = nn.BatchNorm2d(out_c) - - def forward(self, x): - return F.relu(self.bn(self.conv(x))) - - -class ConvLayers(nn.Module): - """Preprocessing layers.""" - - def __init__(self): - super(ConvLayers, self).__init__() - self.conv1 = ConvBlock(3, 48, k=3, s=1, p=1) - self.conv2 = ConvBlock(48, 96, k=3, s=1, p=1) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - - def forward(self, x): - x = self.conv1(x) - x = self.conv2(x) - x = self.maxpool(x) - return x - - -class MultiScaleA(nn.Module): - """Multi-scale stream layer A (Sec.3.1)""" - - def __init__(self): - super(MultiScaleA, self).__init__() - self.stream1 = nn.Sequential( - ConvBlock(96, 96, k=1, s=1, p=0), - ConvBlock(96, 24, k=3, s=1, p=1), - ) - self.stream2 = nn.Sequential( - nn.AvgPool2d(kernel_size=3, stride=1, padding=1), - ConvBlock(96, 24, k=1, s=1, p=0), - ) - self.stream3 = ConvBlock(96, 24, k=1, s=1, p=0) - self.stream4 = nn.Sequential( - ConvBlock(96, 16, k=1, s=1, p=0), - ConvBlock(16, 24, k=3, s=1, p=1), - ConvBlock(24, 24, k=3, s=1, p=1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - s4 = self.stream4(x) - y = torch.cat([s1, s2, s3, s4], dim=1) - return y - - -class Reduction(nn.Module): - """Reduction layer (Sec.3.1)""" - - def __init__(self): - super(Reduction, self).__init__() - self.stream1 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.stream2 = ConvBlock(96, 96, k=3, s=2, p=1) - self.stream3 = nn.Sequential( - ConvBlock(96, 48, k=1, s=1, p=0), - ConvBlock(48, 56, k=3, s=1, p=1), - ConvBlock(56, 64, k=3, s=2, p=1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - y = torch.cat([s1, s2, s3], dim=1) - return y - - -class MultiScaleB(nn.Module): - """Multi-scale stream layer B (Sec.3.1)""" - - def __init__(self): - super(MultiScaleB, self).__init__() - self.stream1 = nn.Sequential( - nn.AvgPool2d(kernel_size=3, stride=1, padding=1), - ConvBlock(256, 256, k=1, s=1, p=0), - ) - self.stream2 = nn.Sequential( - ConvBlock(256, 64, k=1, s=1, p=0), - ConvBlock(64, 128, k=(1, 3), s=1, p=(0, 1)), - ConvBlock(128, 256, k=(3, 1), s=1, p=(1, 0)), - ) - self.stream3 = ConvBlock(256, 256, k=1, s=1, p=0) - self.stream4 = nn.Sequential( - ConvBlock(256, 64, k=1, s=1, p=0), - ConvBlock(64, 64, k=(1, 3), s=1, p=(0, 1)), - ConvBlock(64, 128, k=(3, 1), s=1, p=(1, 0)), - ConvBlock(128, 128, k=(1, 3), s=1, p=(0, 1)), - ConvBlock(128, 256, k=(3, 1), s=1, p=(1, 0)), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - s4 = self.stream4(x) - return s1, s2, s3, s4 - - -class Fusion(nn.Module): - """Saliency-based learning fusion layer (Sec.3.2)""" - - def __init__(self): - super(Fusion, self).__init__() - self.a1 = nn.Parameter(torch.rand(1, 256, 1, 1)) - self.a2 = nn.Parameter(torch.rand(1, 256, 1, 1)) - self.a3 = nn.Parameter(torch.rand(1, 256, 1, 1)) - self.a4 = nn.Parameter(torch.rand(1, 256, 1, 1)) - - # We add an average pooling layer to reduce the spatial dimension - # of feature maps, which differs from the original paper. - self.avgpool = nn.AvgPool2d(kernel_size=4, stride=4, padding=0) - - def forward(self, x1, x2, x3, x4): - s1 = self.a1.expand_as(x1) * x1 - s2 = self.a2.expand_as(x2) * x2 - s3 = self.a3.expand_as(x3) * x3 - s4 = self.a4.expand_as(x4) * x4 - y = self.avgpool(s1 + s2 + s3 + s4) - return y - - -class MuDeep(nn.Module): - """Multiscale deep neural network. - - Reference: - Qian et al. Multi-scale Deep Learning Architectures - for Person Re-identification. ICCV 2017. - - Public keys: - - ``mudeep``: Multiscale deep neural network. - """ - - def __init__(self, num_classes, loss='softmax', **kwargs): - super(MuDeep, self).__init__() - self.loss = loss - - self.block1 = ConvLayers() - self.block2 = MultiScaleA() - self.block3 = Reduction() - self.block4 = MultiScaleB() - self.block5 = Fusion() - - # Due to this fully connected layer, input image has to be fixed - # in shape, i.e. (3, 256, 128), such that the last convolutional feature - # maps are of shape (256, 16, 8). If input shape is changed, - # the input dimension of this layer has to be changed accordingly. - self.fc = nn.Sequential( - nn.Linear(256 * 16 * 8, 4096), - nn.BatchNorm1d(4096), - nn.ReLU(), - ) - self.classifier = nn.Linear(4096, num_classes) - self.feat_dim = 4096 - - def featuremaps(self, x): - x = self.block1(x) - x = self.block2(x) - x = self.block3(x) - x = self.block4(x) - x = self.block5(*x) - return x - - def forward(self, x): - x = self.featuremaps(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - y = self.classifier(x) - - if not self.training: - return x - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, x - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py deleted file mode 100644 index dbf951174d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py +++ /dev/null @@ -1,1178 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.nn.functional as F -import torch.utils.model_zoo as model_zoo - -__all__ = ['nasnetamobile'] -""" -NASNet Mobile -Thanks to Anastasiia (https://github.com/DagnyT) for the great help, support and motivation! - - ------------------------------------------------------------------------------------- - Architecture | Top-1 Acc | Top-5 Acc | Multiply-Adds | Params (M) ------------------------------------------------------------------------------------- -| NASNet-A (4 @ 1056) | 74.08% | 91.74% | 564 M | 5.3 | ------------------------------------------------------------------------------------- -# References: - - [Learning Transferable Architectures for Scalable Image Recognition] - (https://arxiv.org/abs/1707.07012) -""" -""" -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" - -pretrained_settings = { - 'nasnetamobile': { - 'imagenet': { - # 'url': 'https://github.com/veronikayurchuk/pretrained-models.pytorch/releases/download/v1.0/nasnetmobile-7e03cead.pth.tar', - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/nasnetamobile-7e03cead.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], # resize 256 - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000 - }, - # 'imagenet+background': { - # # 'url': 'http://data.lip6.fr/cadene/pretrainedmodels/nasnetalarge-a1897284.pth', - # 'input_space': 'RGB', - # 'input_size': [3, 224, 224], # resize 256 - # 'input_range': [0, 1], - # 'mean': [0.5, 0.5, 0.5], - # 'std': [0.5, 0.5, 0.5], - # 'num_classes': 1001 - # } - } -} - - -class MaxPoolPad(nn.Module): - - def __init__(self): - super(MaxPoolPad, self).__init__() - self.pad = nn.ZeroPad2d((1, 0, 1, 0)) - self.pool = nn.MaxPool2d(3, stride=2, padding=1) - - def forward(self, x): - x = self.pad(x) - x = self.pool(x) - x = x[:, :, 1:, 1:].contiguous() - return x - - -class AvgPoolPad(nn.Module): - - def __init__(self, stride=2, padding=1): - super(AvgPoolPad, self).__init__() - self.pad = nn.ZeroPad2d((1, 0, 1, 0)) - self.pool = nn.AvgPool2d( - 3, stride=stride, padding=padding, count_include_pad=False - ) - - def forward(self, x): - x = self.pad(x) - x = self.pool(x) - x = x[:, :, 1:, 1:].contiguous() - return x - - -class SeparableConv2d(nn.Module): - - def __init__( - self, - in_channels, - out_channels, - dw_kernel, - dw_stride, - dw_padding, - bias=False - ): - super(SeparableConv2d, self).__init__() - self.depthwise_conv2d = nn.Conv2d( - in_channels, - in_channels, - dw_kernel, - stride=dw_stride, - padding=dw_padding, - bias=bias, - groups=in_channels - ) - self.pointwise_conv2d = nn.Conv2d( - in_channels, out_channels, 1, stride=1, bias=bias - ) - - def forward(self, x): - x = self.depthwise_conv2d(x) - x = self.pointwise_conv2d(x) - return x - - -class BranchSeparables(nn.Module): - - def __init__( - self, - in_channels, - out_channels, - kernel_size, - stride, - padding, - name=None, - bias=False - ): - super(BranchSeparables, self).__init__() - self.relu = nn.ReLU() - self.separable_1 = SeparableConv2d( - in_channels, in_channels, kernel_size, stride, padding, bias=bias - ) - self.bn_sep_1 = nn.BatchNorm2d( - in_channels, eps=0.001, momentum=0.1, affine=True - ) - self.relu1 = nn.ReLU() - self.separable_2 = SeparableConv2d( - in_channels, out_channels, kernel_size, 1, padding, bias=bias - ) - self.bn_sep_2 = nn.BatchNorm2d( - out_channels, eps=0.001, momentum=0.1, affine=True - ) - self.name = name - - def forward(self, x): - x = self.relu(x) - if self.name == 'specific': - x = nn.ZeroPad2d((1, 0, 1, 0))(x) - x = self.separable_1(x) - if self.name == 'specific': - x = x[:, :, 1:, 1:].contiguous() - - x = self.bn_sep_1(x) - x = self.relu1(x) - x = self.separable_2(x) - x = self.bn_sep_2(x) - return x - - -class BranchSeparablesStem(nn.Module): - - def __init__( - self, - in_channels, - out_channels, - kernel_size, - stride, - padding, - bias=False - ): - super(BranchSeparablesStem, self).__init__() - self.relu = nn.ReLU() - self.separable_1 = SeparableConv2d( - in_channels, out_channels, kernel_size, stride, padding, bias=bias - ) - self.bn_sep_1 = nn.BatchNorm2d( - out_channels, eps=0.001, momentum=0.1, affine=True - ) - self.relu1 = nn.ReLU() - self.separable_2 = SeparableConv2d( - out_channels, out_channels, kernel_size, 1, padding, bias=bias - ) - self.bn_sep_2 = nn.BatchNorm2d( - out_channels, eps=0.001, momentum=0.1, affine=True - ) - - def forward(self, x): - x = self.relu(x) - x = self.separable_1(x) - x = self.bn_sep_1(x) - x = self.relu1(x) - x = self.separable_2(x) - x = self.bn_sep_2(x) - return x - - -class BranchSeparablesReduction(BranchSeparables): - - def __init__( - self, - in_channels, - out_channels, - kernel_size, - stride, - padding, - z_padding=1, - bias=False - ): - BranchSeparables.__init__( - self, in_channels, out_channels, kernel_size, stride, padding, bias - ) - self.padding = nn.ZeroPad2d((z_padding, 0, z_padding, 0)) - - def forward(self, x): - x = self.relu(x) - x = self.padding(x) - x = self.separable_1(x) - x = x[:, :, 1:, 1:].contiguous() - x = self.bn_sep_1(x) - x = self.relu1(x) - x = self.separable_2(x) - x = self.bn_sep_2(x) - return x - - -class CellStem0(nn.Module): - - def __init__(self, stem_filters, num_filters=42): - super(CellStem0, self).__init__() - self.num_filters = num_filters - self.stem_filters = stem_filters - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - self.stem_filters, self.num_filters, 1, stride=1, bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - self.num_filters, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.comb_iter_0_left = BranchSeparables( - self.num_filters, self.num_filters, 5, 2, 2 - ) - self.comb_iter_0_right = BranchSeparablesStem( - self.stem_filters, self.num_filters, 7, 2, 3, bias=False - ) - - self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) - self.comb_iter_1_right = BranchSeparablesStem( - self.stem_filters, self.num_filters, 7, 2, 3, bias=False - ) - - self.comb_iter_2_left = nn.AvgPool2d( - 3, stride=2, padding=1, count_include_pad=False - ) - self.comb_iter_2_right = BranchSeparablesStem( - self.stem_filters, self.num_filters, 5, 2, 2, bias=False - ) - - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparables( - self.num_filters, self.num_filters, 3, 1, 1, bias=False - ) - self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) - - def forward(self, x): - x1 = self.conv_1x1(x) - - x_comb_iter_0_left = self.comb_iter_0_left(x1) - x_comb_iter_0_right = self.comb_iter_0_right(x) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x1) - x_comb_iter_1_right = self.comb_iter_1_right(x) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x1) - x_comb_iter_2_right = self.comb_iter_2_right(x) - x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right - - x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) - x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 - - x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) - x_comb_iter_4_right = self.comb_iter_4_right(x1) - x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right - - x_out = torch.cat( - [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 - ) - return x_out - - -class CellStem1(nn.Module): - - def __init__(self, stem_filters, num_filters): - super(CellStem1, self).__init__() - self.num_filters = num_filters - self.stem_filters = stem_filters - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - 2 * self.num_filters, - self.num_filters, - 1, - stride=1, - bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - self.num_filters, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.relu = nn.ReLU() - self.path_1 = nn.Sequential() - self.path_1.add_module( - 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) - ) - self.path_1.add_module( - 'conv', - nn.Conv2d( - self.stem_filters, - self.num_filters // 2, - 1, - stride=1, - bias=False - ) - ) - self.path_2 = nn.ModuleList() - self.path_2.add_module('pad', nn.ZeroPad2d((0, 1, 0, 1))) - self.path_2.add_module( - 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) - ) - self.path_2.add_module( - 'conv', - nn.Conv2d( - self.stem_filters, - self.num_filters // 2, - 1, - stride=1, - bias=False - ) - ) - - self.final_path_bn = nn.BatchNorm2d( - self.num_filters, eps=0.001, momentum=0.1, affine=True - ) - - self.comb_iter_0_left = BranchSeparables( - self.num_filters, - self.num_filters, - 5, - 2, - 2, - name='specific', - bias=False - ) - self.comb_iter_0_right = BranchSeparables( - self.num_filters, - self.num_filters, - 7, - 2, - 3, - name='specific', - bias=False - ) - - # self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) - self.comb_iter_1_left = MaxPoolPad() - self.comb_iter_1_right = BranchSeparables( - self.num_filters, - self.num_filters, - 7, - 2, - 3, - name='specific', - bias=False - ) - - # self.comb_iter_2_left = nn.AvgPool2d(3, stride=2, padding=1, count_include_pad=False) - self.comb_iter_2_left = AvgPoolPad() - self.comb_iter_2_right = BranchSeparables( - self.num_filters, - self.num_filters, - 5, - 2, - 2, - name='specific', - bias=False - ) - - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparables( - self.num_filters, - self.num_filters, - 3, - 1, - 1, - name='specific', - bias=False - ) - # self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) - self.comb_iter_4_right = MaxPoolPad() - - def forward(self, x_conv0, x_stem_0): - x_left = self.conv_1x1(x_stem_0) - - x_relu = self.relu(x_conv0) - # path 1 - x_path1 = self.path_1(x_relu) - # path 2 - x_path2 = self.path_2.pad(x_relu) - x_path2 = x_path2[:, :, 1:, 1:] - x_path2 = self.path_2.avgpool(x_path2) - x_path2 = self.path_2.conv(x_path2) - # final path - x_right = self.final_path_bn(torch.cat([x_path1, x_path2], 1)) - - x_comb_iter_0_left = self.comb_iter_0_left(x_left) - x_comb_iter_0_right = self.comb_iter_0_right(x_right) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x_left) - x_comb_iter_1_right = self.comb_iter_1_right(x_right) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x_left) - x_comb_iter_2_right = self.comb_iter_2_right(x_right) - x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right - - x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) - x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 - - x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) - x_comb_iter_4_right = self.comb_iter_4_right(x_left) - x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right - - x_out = torch.cat( - [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 - ) - return x_out - - -class FirstCell(nn.Module): - - def __init__( - self, in_channels_left, out_channels_left, in_channels_right, - out_channels_right - ): - super(FirstCell, self).__init__() - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_right, out_channels_right, 1, stride=1, bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_right, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.relu = nn.ReLU() - self.path_1 = nn.Sequential() - self.path_1.add_module( - 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) - ) - self.path_1.add_module( - 'conv', - nn.Conv2d( - in_channels_left, out_channels_left, 1, stride=1, bias=False - ) - ) - self.path_2 = nn.ModuleList() - self.path_2.add_module('pad', nn.ZeroPad2d((0, 1, 0, 1))) - self.path_2.add_module( - 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) - ) - self.path_2.add_module( - 'conv', - nn.Conv2d( - in_channels_left, out_channels_left, 1, stride=1, bias=False - ) - ) - - self.final_path_bn = nn.BatchNorm2d( - out_channels_left * 2, eps=0.001, momentum=0.1, affine=True - ) - - self.comb_iter_0_left = BranchSeparables( - out_channels_right, out_channels_right, 5, 1, 2, bias=False - ) - self.comb_iter_0_right = BranchSeparables( - out_channels_right, out_channels_right, 3, 1, 1, bias=False - ) - - self.comb_iter_1_left = BranchSeparables( - out_channels_right, out_channels_right, 5, 1, 2, bias=False - ) - self.comb_iter_1_right = BranchSeparables( - out_channels_right, out_channels_right, 3, 1, 1, bias=False - ) - - self.comb_iter_2_left = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_3_left = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparables( - out_channels_right, out_channels_right, 3, 1, 1, bias=False - ) - - def forward(self, x, x_prev): - x_relu = self.relu(x_prev) - # path 1 - x_path1 = self.path_1(x_relu) - # path 2 - x_path2 = self.path_2.pad(x_relu) - x_path2 = x_path2[:, :, 1:, 1:] - x_path2 = self.path_2.avgpool(x_path2) - x_path2 = self.path_2.conv(x_path2) - # final path - x_left = self.final_path_bn(torch.cat([x_path1, x_path2], 1)) - - x_right = self.conv_1x1(x) - - x_comb_iter_0_left = self.comb_iter_0_left(x_right) - x_comb_iter_0_right = self.comb_iter_0_right(x_left) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x_left) - x_comb_iter_1_right = self.comb_iter_1_right(x_left) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x_right) - x_comb_iter_2 = x_comb_iter_2_left + x_left - - x_comb_iter_3_left = self.comb_iter_3_left(x_left) - x_comb_iter_3_right = self.comb_iter_3_right(x_left) - x_comb_iter_3 = x_comb_iter_3_left + x_comb_iter_3_right - - x_comb_iter_4_left = self.comb_iter_4_left(x_right) - x_comb_iter_4 = x_comb_iter_4_left + x_right - - x_out = torch.cat( - [ - x_left, x_comb_iter_0, x_comb_iter_1, x_comb_iter_2, - x_comb_iter_3, x_comb_iter_4 - ], 1 - ) - return x_out - - -class NormalCell(nn.Module): - - def __init__( - self, in_channels_left, out_channels_left, in_channels_right, - out_channels_right - ): - super(NormalCell, self).__init__() - self.conv_prev_1x1 = nn.Sequential() - self.conv_prev_1x1.add_module('relu', nn.ReLU()) - self.conv_prev_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_left, out_channels_left, 1, stride=1, bias=False - ) - ) - self.conv_prev_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_left, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_right, out_channels_right, 1, stride=1, bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_right, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.comb_iter_0_left = BranchSeparables( - out_channels_right, out_channels_right, 5, 1, 2, bias=False - ) - self.comb_iter_0_right = BranchSeparables( - out_channels_left, out_channels_left, 3, 1, 1, bias=False - ) - - self.comb_iter_1_left = BranchSeparables( - out_channels_left, out_channels_left, 5, 1, 2, bias=False - ) - self.comb_iter_1_right = BranchSeparables( - out_channels_left, out_channels_left, 3, 1, 1, bias=False - ) - - self.comb_iter_2_left = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_3_left = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparables( - out_channels_right, out_channels_right, 3, 1, 1, bias=False - ) - - def forward(self, x, x_prev): - x_left = self.conv_prev_1x1(x_prev) - x_right = self.conv_1x1(x) - - x_comb_iter_0_left = self.comb_iter_0_left(x_right) - x_comb_iter_0_right = self.comb_iter_0_right(x_left) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x_left) - x_comb_iter_1_right = self.comb_iter_1_right(x_left) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x_right) - x_comb_iter_2 = x_comb_iter_2_left + x_left - - x_comb_iter_3_left = self.comb_iter_3_left(x_left) - x_comb_iter_3_right = self.comb_iter_3_right(x_left) - x_comb_iter_3 = x_comb_iter_3_left + x_comb_iter_3_right - - x_comb_iter_4_left = self.comb_iter_4_left(x_right) - x_comb_iter_4 = x_comb_iter_4_left + x_right - - x_out = torch.cat( - [ - x_left, x_comb_iter_0, x_comb_iter_1, x_comb_iter_2, - x_comb_iter_3, x_comb_iter_4 - ], 1 - ) - return x_out - - -class ReductionCell0(nn.Module): - - def __init__( - self, in_channels_left, out_channels_left, in_channels_right, - out_channels_right - ): - super(ReductionCell0, self).__init__() - self.conv_prev_1x1 = nn.Sequential() - self.conv_prev_1x1.add_module('relu', nn.ReLU()) - self.conv_prev_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_left, out_channels_left, 1, stride=1, bias=False - ) - ) - self.conv_prev_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_left, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_right, out_channels_right, 1, stride=1, bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_right, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.comb_iter_0_left = BranchSeparablesReduction( - out_channels_right, out_channels_right, 5, 2, 2, bias=False - ) - self.comb_iter_0_right = BranchSeparablesReduction( - out_channels_right, out_channels_right, 7, 2, 3, bias=False - ) - - self.comb_iter_1_left = MaxPoolPad() - self.comb_iter_1_right = BranchSeparablesReduction( - out_channels_right, out_channels_right, 7, 2, 3, bias=False - ) - - self.comb_iter_2_left = AvgPoolPad() - self.comb_iter_2_right = BranchSeparablesReduction( - out_channels_right, out_channels_right, 5, 2, 2, bias=False - ) - - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparablesReduction( - out_channels_right, out_channels_right, 3, 1, 1, bias=False - ) - self.comb_iter_4_right = MaxPoolPad() - - def forward(self, x, x_prev): - x_left = self.conv_prev_1x1(x_prev) - x_right = self.conv_1x1(x) - - x_comb_iter_0_left = self.comb_iter_0_left(x_right) - x_comb_iter_0_right = self.comb_iter_0_right(x_left) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x_right) - x_comb_iter_1_right = self.comb_iter_1_right(x_left) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x_right) - x_comb_iter_2_right = self.comb_iter_2_right(x_left) - x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right - - x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) - x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 - - x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) - x_comb_iter_4_right = self.comb_iter_4_right(x_right) - x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right - - x_out = torch.cat( - [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 - ) - return x_out - - -class ReductionCell1(nn.Module): - - def __init__( - self, in_channels_left, out_channels_left, in_channels_right, - out_channels_right - ): - super(ReductionCell1, self).__init__() - self.conv_prev_1x1 = nn.Sequential() - self.conv_prev_1x1.add_module('relu', nn.ReLU()) - self.conv_prev_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_left, out_channels_left, 1, stride=1, bias=False - ) - ) - self.conv_prev_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_left, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.conv_1x1 = nn.Sequential() - self.conv_1x1.add_module('relu', nn.ReLU()) - self.conv_1x1.add_module( - 'conv', - nn.Conv2d( - in_channels_right, out_channels_right, 1, stride=1, bias=False - ) - ) - self.conv_1x1.add_module( - 'bn', - nn.BatchNorm2d( - out_channels_right, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.comb_iter_0_left = BranchSeparables( - out_channels_right, - out_channels_right, - 5, - 2, - 2, - name='specific', - bias=False - ) - self.comb_iter_0_right = BranchSeparables( - out_channels_right, - out_channels_right, - 7, - 2, - 3, - name='specific', - bias=False - ) - - # self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) - self.comb_iter_1_left = MaxPoolPad() - self.comb_iter_1_right = BranchSeparables( - out_channels_right, - out_channels_right, - 7, - 2, - 3, - name='specific', - bias=False - ) - - # self.comb_iter_2_left = nn.AvgPool2d(3, stride=2, padding=1, count_include_pad=False) - self.comb_iter_2_left = AvgPoolPad() - self.comb_iter_2_right = BranchSeparables( - out_channels_right, - out_channels_right, - 5, - 2, - 2, - name='specific', - bias=False - ) - - self.comb_iter_3_right = nn.AvgPool2d( - 3, stride=1, padding=1, count_include_pad=False - ) - - self.comb_iter_4_left = BranchSeparables( - out_channels_right, - out_channels_right, - 3, - 1, - 1, - name='specific', - bias=False - ) - # self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) - self.comb_iter_4_right = MaxPoolPad() - - def forward(self, x, x_prev): - x_left = self.conv_prev_1x1(x_prev) - x_right = self.conv_1x1(x) - - x_comb_iter_0_left = self.comb_iter_0_left(x_right) - x_comb_iter_0_right = self.comb_iter_0_right(x_left) - x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right - - x_comb_iter_1_left = self.comb_iter_1_left(x_right) - x_comb_iter_1_right = self.comb_iter_1_right(x_left) - x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right - - x_comb_iter_2_left = self.comb_iter_2_left(x_right) - x_comb_iter_2_right = self.comb_iter_2_right(x_left) - x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right - - x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) - x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 - - x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) - x_comb_iter_4_right = self.comb_iter_4_right(x_right) - x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right - - x_out = torch.cat( - [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 - ) - return x_out - - -class NASNetAMobile(nn.Module): - """Neural Architecture Search (NAS). - - Reference: - Zoph et al. Learning Transferable Architectures - for Scalable Image Recognition. CVPR 2018. - - Public keys: - - ``nasnetamobile``: NASNet-A Mobile. - """ - - def __init__( - self, - num_classes, - loss, - stem_filters=32, - penultimate_filters=1056, - filters_multiplier=2, - **kwargs - ): - super(NASNetAMobile, self).__init__() - self.stem_filters = stem_filters - self.penultimate_filters = penultimate_filters - self.filters_multiplier = filters_multiplier - self.loss = loss - - filters = self.penultimate_filters // 24 - # 24 is default value for the architecture - - self.conv0 = nn.Sequential() - self.conv0.add_module( - 'conv', - nn.Conv2d( - in_channels=3, - out_channels=self.stem_filters, - kernel_size=3, - padding=0, - stride=2, - bias=False - ) - ) - self.conv0.add_module( - 'bn', - nn.BatchNorm2d( - self.stem_filters, eps=0.001, momentum=0.1, affine=True - ) - ) - - self.cell_stem_0 = CellStem0( - self.stem_filters, num_filters=filters // (filters_multiplier**2) - ) - self.cell_stem_1 = CellStem1( - self.stem_filters, num_filters=filters // filters_multiplier - ) - - self.cell_0 = FirstCell( - in_channels_left=filters, - out_channels_left=filters // 2, # 1, 0.5 - in_channels_right=2 * filters, - out_channels_right=filters - ) # 2, 1 - self.cell_1 = NormalCell( - in_channels_left=2 * filters, - out_channels_left=filters, # 2, 1 - in_channels_right=6 * filters, - out_channels_right=filters - ) # 6, 1 - self.cell_2 = NormalCell( - in_channels_left=6 * filters, - out_channels_left=filters, # 6, 1 - in_channels_right=6 * filters, - out_channels_right=filters - ) # 6, 1 - self.cell_3 = NormalCell( - in_channels_left=6 * filters, - out_channels_left=filters, # 6, 1 - in_channels_right=6 * filters, - out_channels_right=filters - ) # 6, 1 - - self.reduction_cell_0 = ReductionCell0( - in_channels_left=6 * filters, - out_channels_left=2 * filters, # 6, 2 - in_channels_right=6 * filters, - out_channels_right=2 * filters - ) # 6, 2 - - self.cell_6 = FirstCell( - in_channels_left=6 * filters, - out_channels_left=filters, # 6, 1 - in_channels_right=8 * filters, - out_channels_right=2 * filters - ) # 8, 2 - self.cell_7 = NormalCell( - in_channels_left=8 * filters, - out_channels_left=2 * filters, # 8, 2 - in_channels_right=12 * filters, - out_channels_right=2 * filters - ) # 12, 2 - self.cell_8 = NormalCell( - in_channels_left=12 * filters, - out_channels_left=2 * filters, # 12, 2 - in_channels_right=12 * filters, - out_channels_right=2 * filters - ) # 12, 2 - self.cell_9 = NormalCell( - in_channels_left=12 * filters, - out_channels_left=2 * filters, # 12, 2 - in_channels_right=12 * filters, - out_channels_right=2 * filters - ) # 12, 2 - - self.reduction_cell_1 = ReductionCell1( - in_channels_left=12 * filters, - out_channels_left=4 * filters, # 12, 4 - in_channels_right=12 * filters, - out_channels_right=4 * filters - ) # 12, 4 - - self.cell_12 = FirstCell( - in_channels_left=12 * filters, - out_channels_left=2 * filters, # 12, 2 - in_channels_right=16 * filters, - out_channels_right=4 * filters - ) # 16, 4 - self.cell_13 = NormalCell( - in_channels_left=16 * filters, - out_channels_left=4 * filters, # 16, 4 - in_channels_right=24 * filters, - out_channels_right=4 * filters - ) # 24, 4 - self.cell_14 = NormalCell( - in_channels_left=24 * filters, - out_channels_left=4 * filters, # 24, 4 - in_channels_right=24 * filters, - out_channels_right=4 * filters - ) # 24, 4 - self.cell_15 = NormalCell( - in_channels_left=24 * filters, - out_channels_left=4 * filters, # 24, 4 - in_channels_right=24 * filters, - out_channels_right=4 * filters - ) # 24, 4 - - self.relu = nn.ReLU() - self.dropout = nn.Dropout() - self.classifier = nn.Linear(24 * filters, num_classes) - - self._init_params() - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def features(self, input): - x_conv0 = self.conv0(input) - x_stem_0 = self.cell_stem_0(x_conv0) - x_stem_1 = self.cell_stem_1(x_conv0, x_stem_0) - - x_cell_0 = self.cell_0(x_stem_1, x_stem_0) - x_cell_1 = self.cell_1(x_cell_0, x_stem_1) - x_cell_2 = self.cell_2(x_cell_1, x_cell_0) - x_cell_3 = self.cell_3(x_cell_2, x_cell_1) - - x_reduction_cell_0 = self.reduction_cell_0(x_cell_3, x_cell_2) - - x_cell_6 = self.cell_6(x_reduction_cell_0, x_cell_3) - x_cell_7 = self.cell_7(x_cell_6, x_reduction_cell_0) - x_cell_8 = self.cell_8(x_cell_7, x_cell_6) - x_cell_9 = self.cell_9(x_cell_8, x_cell_7) - - x_reduction_cell_1 = self.reduction_cell_1(x_cell_9, x_cell_8) - - x_cell_12 = self.cell_12(x_reduction_cell_1, x_cell_9) - x_cell_13 = self.cell_13(x_cell_12, x_reduction_cell_1) - x_cell_14 = self.cell_14(x_cell_13, x_cell_12) - x_cell_15 = self.cell_15(x_cell_14, x_cell_13) - - x_cell_15 = self.relu(x_cell_15) - x_cell_15 = F.avg_pool2d( - x_cell_15, - x_cell_15.size()[2:] - ) # global average pool - x_cell_15 = x_cell_15.view(x_cell_15.size(0), -1) - x_cell_15 = self.dropout(x_cell_15) - - return x_cell_15 - - def forward(self, input): - v = self.features(input) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def nasnetamobile(num_classes, loss='softmax', pretrained=True, **kwargs): - model = NASNetAMobile(num_classes, loss, **kwargs) - if pretrained: - model_url = pretrained_settings['nasnetamobile']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py deleted file mode 100644 index f831d02648..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py +++ /dev/null @@ -1,645 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import warnings -import torch -from torch import nn -from torch.nn import functional as F - -__all__ = [ - 'osnet_x1_0', 'osnet_x0_75', 'osnet_x0_5', 'osnet_x0_25', 'osnet_ibn_x1_0' -] - -pretrained_urls = { - 'osnet_x1_0': - 'https://drive.google.com/uc?id=1LaG1EJpHrxdAxKnSCJ_i0u-nbxSAeiFY', - 'osnet_x0_75': - 'https://drive.google.com/uc?id=1uwA9fElHOk3ZogwbeY5GkLI6QPTX70Hq', - 'osnet_x0_5': - 'https://drive.google.com/uc?id=16DGLbZukvVYgINws8u8deSaOqjybZ83i', - 'osnet_x0_25': - 'https://drive.google.com/uc?id=1rb8UN5ZzPKRc_xvtHlyDh-cSz88YX9hs', - 'osnet_ibn_x1_0': - 'https://drive.google.com/uc?id=1sr90V6irlYYDd4_4ISU2iruoRG8J__6l' -} - - -########## -# Basic layers -########## -class ConvLayer(nn.Module): - """Convolution layer (conv + bn + relu).""" - - def __init__( - self, - in_channels, - out_channels, - kernel_size, - stride=1, - padding=0, - groups=1, - IN=False - ): - super(ConvLayer, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - kernel_size, - stride=stride, - padding=padding, - bias=False, - groups=groups - ) - if IN: - self.bn = nn.InstanceNorm2d(out_channels, affine=True) - else: - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Conv1x1(nn.Module): - """1x1 convolution + bn + relu.""" - - def __init__(self, in_channels, out_channels, stride=1, groups=1): - super(Conv1x1, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - 1, - stride=stride, - padding=0, - bias=False, - groups=groups - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Conv1x1Linear(nn.Module): - """1x1 convolution + bn (w/o non-linearity).""" - - def __init__(self, in_channels, out_channels, stride=1): - super(Conv1x1Linear, self).__init__() - self.conv = nn.Conv2d( - in_channels, out_channels, 1, stride=stride, padding=0, bias=False - ) - self.bn = nn.BatchNorm2d(out_channels) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - return x - - -class Conv3x3(nn.Module): - """3x3 convolution + bn + relu.""" - - def __init__(self, in_channels, out_channels, stride=1, groups=1): - super(Conv3x3, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - 3, - stride=stride, - padding=1, - bias=False, - groups=groups - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class LightConv3x3(nn.Module): - """Lightweight 3x3 convolution. - - 1x1 (linear) + dw 3x3 (nonlinear). - """ - - def __init__(self, in_channels, out_channels): - super(LightConv3x3, self).__init__() - self.conv1 = nn.Conv2d( - in_channels, out_channels, 1, stride=1, padding=0, bias=False - ) - self.conv2 = nn.Conv2d( - out_channels, - out_channels, - 3, - stride=1, - padding=1, - bias=False, - groups=out_channels - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv1(x) - x = self.conv2(x) - x = self.bn(x) - x = self.relu(x) - return x - - -########## -# Building blocks for omni-scale feature learning -########## -class ChannelGate(nn.Module): - """A mini-network that generates channel-wise gates conditioned on input tensor.""" - - def __init__( - self, - in_channels, - num_gates=None, - return_gates=False, - gate_activation='sigmoid', - reduction=16, - layer_norm=False - ): - super(ChannelGate, self).__init__() - if num_gates is None: - num_gates = in_channels - self.return_gates = return_gates - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc1 = nn.Conv2d( - in_channels, - in_channels // reduction, - kernel_size=1, - bias=True, - padding=0 - ) - self.norm1 = None - if layer_norm: - self.norm1 = nn.LayerNorm((in_channels // reduction, 1, 1)) - self.relu = nn.ReLU(inplace=True) - self.fc2 = nn.Conv2d( - in_channels // reduction, - num_gates, - kernel_size=1, - bias=True, - padding=0 - ) - if gate_activation == 'sigmoid': - self.gate_activation = nn.Sigmoid() - elif gate_activation == 'relu': - self.gate_activation = nn.ReLU(inplace=True) - elif gate_activation == 'linear': - self.gate_activation = None - else: - raise RuntimeError( - "Unknown gate activation: {}".format(gate_activation) - ) - - def forward(self, x): - input = x - x = self.global_avgpool(x) - x = self.fc1(x) - if self.norm1 is not None: - x = self.norm1(x) - x = self.relu(x) - x = self.fc2(x) - if self.gate_activation is not None: - x = self.gate_activation(x) - if self.return_gates: - return x - return input * x - - -class OSBlock(nn.Module): - """Omni-scale feature learning block.""" - - def __init__( - self, - in_channels, - out_channels, - IN=False, - bottleneck_reduction=4, - **kwargs - ): - super(OSBlock, self).__init__() - mid_channels = out_channels // bottleneck_reduction - self.conv1 = Conv1x1(in_channels, mid_channels) - self.conv2a = LightConv3x3(mid_channels, mid_channels) - self.conv2b = nn.Sequential( - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - ) - self.conv2c = nn.Sequential( - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - ) - self.conv2d = nn.Sequential( - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - LightConv3x3(mid_channels, mid_channels), - ) - self.gate = ChannelGate(mid_channels) - self.conv3 = Conv1x1Linear(mid_channels, out_channels) - self.downsample = None - if in_channels != out_channels: - self.downsample = Conv1x1Linear(in_channels, out_channels) - self.IN = None - if IN: - self.IN = nn.InstanceNorm2d(out_channels, affine=True) - - def forward(self, x): - identity = x - x1 = self.conv1(x) - x2a = self.conv2a(x1) - x2b = self.conv2b(x1) - x2c = self.conv2c(x1) - x2d = self.conv2d(x1) - x2 = self.gate(x2a) + self.gate(x2b) + self.gate(x2c) + self.gate(x2d) - x3 = self.conv3(x2) - if self.downsample is not None: - identity = self.downsample(identity) - out = x3 + identity - if self.IN is not None: - out = self.IN(out) - return F.relu(out) - - -########## -# Network architecture -########## -class OSNet(nn.Module): - """Omni-Scale Network. - - Reference: - - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. - - Zhou et al. Learning Generalisable Omni-Scale Representations - for Person Re-Identification. TPAMI, 2021. - """ - - def __init__( - self, - num_classes, - blocks, - layers, - channels, - feature_dim=512, - loss='softmax', - IN=False, - **kwargs - ): - super(OSNet, self).__init__() - num_blocks = len(blocks) - assert num_blocks == len(layers) - assert num_blocks == len(channels) - 1 - self.loss = loss - self.feature_dim = feature_dim - - # convolutional backbone - self.conv1 = ConvLayer(3, channels[0], 7, stride=2, padding=3, IN=IN) - self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) - self.conv2 = self._make_layer( - blocks[0], - layers[0], - channels[0], - channels[1], - reduce_spatial_size=True, - IN=IN - ) - self.conv3 = self._make_layer( - blocks[1], - layers[1], - channels[1], - channels[2], - reduce_spatial_size=True - ) - self.conv4 = self._make_layer( - blocks[2], - layers[2], - channels[2], - channels[3], - reduce_spatial_size=False - ) - self.conv5 = Conv1x1(channels[3], channels[3]) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - # fully connected layer - self.fc = self._construct_fc_layer( - self.feature_dim, channels[3], dropout_p=None - ) - # identity classification layer - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _make_layer( - self, - block, - layer, - in_channels, - out_channels, - reduce_spatial_size, - IN=False - ): - layers = [] - - layers.append(block(in_channels, out_channels, IN=IN)) - for i in range(1, layer): - layers.append(block(out_channels, out_channels, IN=IN)) - - if reduce_spatial_size: - layers.append( - nn.Sequential( - Conv1x1(out_channels, out_channels), - nn.AvgPool2d(2, stride=2) - ) - ) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - if fc_dims is None or fc_dims < 0: - self.feature_dim = input_dim - return None - - if isinstance(fc_dims, int): - fc_dims = [fc_dims] - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.maxpool(x) - x = self.conv2(x) - x = self.conv3(x) - x = self.conv4(x) - x = self.conv5(x) - return x - - def forward(self, x, return_featuremaps=False): - x = self.featuremaps(x) - if return_featuremaps: - return x - v = self.global_avgpool(x) - v = v.view(v.size(0), -1) - if self.fc is not None: - v = self.fc(v) - if not self.training: - return v - y = self.classifier(v) - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, key=''): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - import os - import errno - import gdown - from collections import OrderedDict - - def _get_torch_home(): - ENV_TORCH_HOME = 'TORCH_HOME' - ENV_XDG_CACHE_HOME = 'XDG_CACHE_HOME' - DEFAULT_CACHE_DIR = '~/.cache' - torch_home = os.path.expanduser( - os.getenv( - ENV_TORCH_HOME, - os.path.join( - os.getenv(ENV_XDG_CACHE_HOME, DEFAULT_CACHE_DIR), 'torch' - ) - ) - ) - return torch_home - - torch_home = _get_torch_home() - model_dir = os.path.join(torch_home, 'checkpoints') - try: - os.makedirs(model_dir) - except OSError as e: - if e.errno == errno.EEXIST: - # Directory already exists, ignore. - pass - else: - # Unexpected OSError, re-raise. - raise - filename = key + '_imagenet.pth' - cached_file = os.path.join(model_dir, filename) - - if not os.path.exists(cached_file): - gdown.download(pretrained_urls[key], cached_file, quiet=False) - - state_dict = torch.load(cached_file) - model_dict = model.state_dict() - new_state_dict = OrderedDict() - matched_layers, discarded_layers = [], [] - - for k, v in state_dict.items(): - if k.startswith('module.'): - k = k[7:] # discard module. - - if k in model_dict and model_dict[k].size() == v.size(): - new_state_dict[k] = v - matched_layers.append(k) - else: - discarded_layers.append(k) - - model_dict.update(new_state_dict) - model.load_state_dict(model_dict) - - if len(matched_layers) == 0: - warnings.warn( - 'The pretrained weights from "{}" cannot be loaded, ' - 'please check the key names manually ' - '(** ignored and continue **)'.format(cached_file) - ) - else: - print( - 'Successfully loaded imagenet pretrained weights from "{}"'. - format(cached_file) - ) - if len(discarded_layers) > 0: - print( - '** The following layers are discarded ' - 'due to unmatched keys or layer size: {}'. - format(discarded_layers) - ) - - -########## -# Instantiation -########## -def osnet_x1_0(num_classes=1000, pretrained=True, loss='softmax', **kwargs): - # standard size (width x1.0) - model = OSNet( - num_classes, - blocks=[OSBlock, OSBlock, OSBlock], - layers=[2, 2, 2], - channels=[64, 256, 384, 512], - loss=loss, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_x1_0') - return model - - -def osnet_x0_75(num_classes=1000, pretrained=True, loss='softmax', **kwargs): - # medium size (width x0.75) - model = OSNet( - num_classes, - blocks=[OSBlock, OSBlock, OSBlock], - layers=[2, 2, 2], - channels=[48, 192, 288, 384], - loss=loss, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_x0_75') - return model - - -def osnet_x0_5(num_classes=1000, pretrained=True, loss='softmax', **kwargs): - # tiny size (width x0.5) - model = OSNet( - num_classes, - blocks=[OSBlock, OSBlock, OSBlock], - layers=[2, 2, 2], - channels=[32, 128, 192, 256], - loss=loss, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_x0_5') - return model - - -def osnet_x0_25(num_classes=1000, pretrained=True, loss='softmax', **kwargs): - # very tiny size (width x0.25) - model = OSNet( - num_classes, - blocks=[OSBlock, OSBlock, OSBlock], - layers=[2, 2, 2], - channels=[16, 64, 96, 128], - loss=loss, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_x0_25') - return model - - -def osnet_ibn_x1_0( - num_classes=1000, pretrained=True, loss='softmax', **kwargs -): - # standard size (width x1.0) + IBN layer - # Ref: Pan et al. Two at Once: Enhancing Learning and Generalization Capacities via IBN-Net. ECCV, 2018. - model = OSNet( - num_classes, - blocks=[OSBlock, OSBlock, OSBlock], - layers=[2, 2, 2], - channels=[64, 256, 384, 512], - loss=loss, - IN=True, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_ibn_x1_0') - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py deleted file mode 100644 index 3f71e7a528..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py +++ /dev/null @@ -1,588 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import warnings -import torch -from torch import nn -from torch.nn import functional as F - -__all__ = ['osnet_ain_x1_0'] - -pretrained_urls = { - 'osnet_ain_x1_0': - 'https://drive.google.com/uc?id=1-CaioD9NaqbHK_kzSMW8VE4_3KcsRjEo' -} - - -########## -# Basic layers -########## -class ConvLayer(nn.Module): - """Convolution layer (conv + bn + relu).""" - - def __init__( - self, - in_channels, - out_channels, - kernel_size, - stride=1, - padding=0, - groups=1, - IN=False - ): - super(ConvLayer, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - kernel_size, - stride=stride, - padding=padding, - bias=False, - groups=groups - ) - if IN: - self.bn = nn.InstanceNorm2d(out_channels, affine=True) - else: - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU() - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - return self.relu(x) - - -class Conv1x1(nn.Module): - """1x1 convolution + bn + relu.""" - - def __init__(self, in_channels, out_channels, stride=1, groups=1): - super(Conv1x1, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - 1, - stride=stride, - padding=0, - bias=False, - groups=groups - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU() - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - return self.relu(x) - - -class Conv1x1Linear(nn.Module): - """1x1 convolution + bn (w/o non-linearity).""" - - def __init__(self, in_channels, out_channels, stride=1, bn=True): - super(Conv1x1Linear, self).__init__() - self.conv = nn.Conv2d( - in_channels, out_channels, 1, stride=stride, padding=0, bias=False - ) - self.bn = None - if bn: - self.bn = nn.BatchNorm2d(out_channels) - - def forward(self, x): - x = self.conv(x) - if self.bn is not None: - x = self.bn(x) - return x - - -class Conv3x3(nn.Module): - """3x3 convolution + bn + relu.""" - - def __init__(self, in_channels, out_channels, stride=1, groups=1): - super(Conv3x3, self).__init__() - self.conv = nn.Conv2d( - in_channels, - out_channels, - 3, - stride=stride, - padding=1, - bias=False, - groups=groups - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU() - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - return self.relu(x) - - -class LightConv3x3(nn.Module): - """Lightweight 3x3 convolution. - - 1x1 (linear) + dw 3x3 (nonlinear). - """ - - def __init__(self, in_channels, out_channels): - super(LightConv3x3, self).__init__() - self.conv1 = nn.Conv2d( - in_channels, out_channels, 1, stride=1, padding=0, bias=False - ) - self.conv2 = nn.Conv2d( - out_channels, - out_channels, - 3, - stride=1, - padding=1, - bias=False, - groups=out_channels - ) - self.bn = nn.BatchNorm2d(out_channels) - self.relu = nn.ReLU() - - def forward(self, x): - x = self.conv1(x) - x = self.conv2(x) - x = self.bn(x) - return self.relu(x) - - -class LightConvStream(nn.Module): - """Lightweight convolution stream.""" - - def __init__(self, in_channels, out_channels, depth): - super(LightConvStream, self).__init__() - assert depth >= 1, 'depth must be equal to or larger than 1, but got {}'.format( - depth - ) - layers = [] - layers += [LightConv3x3(in_channels, out_channels)] - for i in range(depth - 1): - layers += [LightConv3x3(out_channels, out_channels)] - self.layers = nn.Sequential(*layers) - - def forward(self, x): - return self.layers(x) - - -########## -# Building blocks for omni-scale feature learning -########## -class ChannelGate(nn.Module): - """A mini-network that generates channel-wise gates conditioned on input tensor.""" - - def __init__( - self, - in_channels, - num_gates=None, - return_gates=False, - gate_activation='sigmoid', - reduction=16, - layer_norm=False - ): - super(ChannelGate, self).__init__() - if num_gates is None: - num_gates = in_channels - self.return_gates = return_gates - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc1 = nn.Conv2d( - in_channels, - in_channels // reduction, - kernel_size=1, - bias=True, - padding=0 - ) - self.norm1 = None - if layer_norm: - self.norm1 = nn.LayerNorm((in_channels // reduction, 1, 1)) - self.relu = nn.ReLU() - self.fc2 = nn.Conv2d( - in_channels // reduction, - num_gates, - kernel_size=1, - bias=True, - padding=0 - ) - if gate_activation == 'sigmoid': - self.gate_activation = nn.Sigmoid() - elif gate_activation == 'relu': - self.gate_activation = nn.ReLU() - elif gate_activation == 'linear': - self.gate_activation = None - else: - raise RuntimeError( - "Unknown gate activation: {}".format(gate_activation) - ) - - def forward(self, x): - input = x - x = self.global_avgpool(x) - x = self.fc1(x) - if self.norm1 is not None: - x = self.norm1(x) - x = self.relu(x) - x = self.fc2(x) - if self.gate_activation is not None: - x = self.gate_activation(x) - if self.return_gates: - return x - return input * x - - -class OSBlock(nn.Module): - """Omni-scale feature learning block.""" - - def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs): - super(OSBlock, self).__init__() - assert T >= 1 - assert out_channels >= reduction and out_channels % reduction == 0 - mid_channels = out_channels // reduction - - self.conv1 = Conv1x1(in_channels, mid_channels) - self.conv2 = nn.ModuleList() - for t in range(1, T + 1): - self.conv2 += [LightConvStream(mid_channels, mid_channels, t)] - self.gate = ChannelGate(mid_channels) - self.conv3 = Conv1x1Linear(mid_channels, out_channels) - self.downsample = None - if in_channels != out_channels: - self.downsample = Conv1x1Linear(in_channels, out_channels) - - def forward(self, x): - identity = x - x1 = self.conv1(x) - x2 = 0 - for conv2_t in self.conv2: - x2_t = conv2_t(x1) - x2 = x2 + self.gate(x2_t) - x3 = self.conv3(x2) - if self.downsample is not None: - identity = self.downsample(identity) - out = x3 + identity - return F.relu(out) - - -class OSBlockINin(nn.Module): - """Omni-scale feature learning block with instance normalization.""" - - def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs): - super(OSBlockINin, self).__init__() - assert T >= 1 - assert out_channels >= reduction and out_channels % reduction == 0 - mid_channels = out_channels // reduction - - self.conv1 = Conv1x1(in_channels, mid_channels) - self.conv2 = nn.ModuleList() - for t in range(1, T + 1): - self.conv2 += [LightConvStream(mid_channels, mid_channels, t)] - self.gate = ChannelGate(mid_channels) - self.conv3 = Conv1x1Linear(mid_channels, out_channels, bn=False) - self.downsample = None - if in_channels != out_channels: - self.downsample = Conv1x1Linear(in_channels, out_channels) - self.IN = nn.InstanceNorm2d(out_channels, affine=True) - - def forward(self, x): - identity = x - x1 = self.conv1(x) - x2 = 0 - for conv2_t in self.conv2: - x2_t = conv2_t(x1) - x2 = x2 + self.gate(x2_t) - x3 = self.conv3(x2) - x3 = self.IN(x3) # IN inside residual - if self.downsample is not None: - identity = self.downsample(identity) - out = x3 + identity - return F.relu(out) - - -########## -# Network architecture -########## -class OSNet(nn.Module): - """Omni-Scale Network. - - Reference: - - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. - - Zhou et al. Learning Generalisable Omni-Scale Representations - for Person Re-Identification. TPAMI, 2021. - """ - - def __init__( - self, - num_classes, - blocks, - layers, - channels, - feature_dim=512, - loss='softmax', - conv1_IN=False, - **kwargs - ): - super(OSNet, self).__init__() - num_blocks = len(blocks) - assert num_blocks == len(layers) - assert num_blocks == len(channels) - 1 - self.loss = loss - self.feature_dim = feature_dim - - # convolutional backbone - self.conv1 = ConvLayer( - 3, channels[0], 7, stride=2, padding=3, IN=conv1_IN - ) - self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) - self.conv2 = self._make_layer( - blocks[0], layers[0], channels[0], channels[1] - ) - self.pool2 = nn.Sequential( - Conv1x1(channels[1], channels[1]), nn.AvgPool2d(2, stride=2) - ) - self.conv3 = self._make_layer( - blocks[1], layers[1], channels[1], channels[2] - ) - self.pool3 = nn.Sequential( - Conv1x1(channels[2], channels[2]), nn.AvgPool2d(2, stride=2) - ) - self.conv4 = self._make_layer( - blocks[2], layers[2], channels[2], channels[3] - ) - self.conv5 = Conv1x1(channels[3], channels[3]) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - # fully connected layer - self.fc = self._construct_fc_layer( - self.feature_dim, channels[3], dropout_p=None - ) - # identity classification layer - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _make_layer(self, blocks, layer, in_channels, out_channels): - layers = [] - layers += [blocks[0](in_channels, out_channels)] - for i in range(1, len(blocks)): - layers += [blocks[i](out_channels, out_channels)] - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - if fc_dims is None or fc_dims < 0: - self.feature_dim = input_dim - return None - - if isinstance(fc_dims, int): - fc_dims = [fc_dims] - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU()) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.InstanceNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.maxpool(x) - x = self.conv2(x) - x = self.pool2(x) - x = self.conv3(x) - x = self.pool3(x) - x = self.conv4(x) - x = self.conv5(x) - return x - - def forward(self, x, return_featuremaps=False): - x = self.featuremaps(x) - if return_featuremaps: - return x - v = self.global_avgpool(x) - v = v.view(v.size(0), -1) - if self.fc is not None: - v = self.fc(v) - if not self.training: - return v - y = self.classifier(v) - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, key=''): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - import os - import errno - import gdown - from collections import OrderedDict - - def _get_torch_home(): - ENV_TORCH_HOME = 'TORCH_HOME' - ENV_XDG_CACHE_HOME = 'XDG_CACHE_HOME' - DEFAULT_CACHE_DIR = '~/.cache' - torch_home = os.path.expanduser( - os.getenv( - ENV_TORCH_HOME, - os.path.join( - os.getenv(ENV_XDG_CACHE_HOME, DEFAULT_CACHE_DIR), 'torch' - ) - ) - ) - return torch_home - - torch_home = _get_torch_home() - model_dir = os.path.join(torch_home, 'checkpoints') - try: - os.makedirs(model_dir) - except OSError as e: - if e.errno == errno.EEXIST: - # Directory already exists, ignore. - pass - else: - # Unexpected OSError, re-raise. - raise - filename = key + '_imagenet.pth' - cached_file = os.path.join(model_dir, filename) - - if not os.path.exists(cached_file): - gdown.download(pretrained_urls[key], cached_file, quiet=False) - - state_dict = torch.load(cached_file) - model_dict = model.state_dict() - new_state_dict = OrderedDict() - matched_layers, discarded_layers = [], [] - - for k, v in state_dict.items(): - if k.startswith('module.'): - k = k[7:] # discard module. - - if k in model_dict and model_dict[k].size() == v.size(): - new_state_dict[k] = v - matched_layers.append(k) - else: - discarded_layers.append(k) - - model_dict.update(new_state_dict) - model.load_state_dict(model_dict) - - if len(matched_layers) == 0: - warnings.warn( - 'The pretrained weights from "{}" cannot be loaded, ' - 'please check the key names manually ' - '(** ignored and continue **)'.format(cached_file) - ) - else: - print( - 'Successfully loaded imagenet pretrained weights from "{}"'. - format(cached_file) - ) - if len(discarded_layers) > 0: - print( - '** The following layers are discarded ' - 'due to unmatched keys or layer size: {}'. - format(discarded_layers) - ) - - -########## -# Instantiation -########## -def osnet_ain_x1_0( - num_classes=1000, pretrained=True, loss='softmax', **kwargs -): - model = OSNet( - num_classes, - blocks=[ - [OSBlockINin, OSBlockINin], [OSBlock, OSBlockINin], - [OSBlockINin, OSBlock] - ], - layers=[2, 2, 2], - channels=[64, 256, 384, 512], - loss=loss, - conv1_IN=True, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, key='osnet_ain_x1_0') - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py deleted file mode 100644 index 8065574baa..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py +++ /dev/null @@ -1,361 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['pcb_p6', 'pcb_p4'] - -model_urls = { - 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', - 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1): - """3x3 convolution with padding""" - return nn.Conv2d( - in_planes, - out_planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(BasicBlock, self).__init__() - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = nn.BatchNorm2d(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = nn.BatchNorm2d(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(Bottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) - self.bn1 = nn.BatchNorm2d(planes) - self.conv2 = nn.Conv2d( - planes, - planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes) - self.conv3 = nn.Conv2d( - planes, planes * self.expansion, kernel_size=1, bias=False - ) - self.bn3 = nn.BatchNorm2d(planes * self.expansion) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class DimReduceLayer(nn.Module): - - def __init__(self, in_channels, out_channels, nonlinear): - super(DimReduceLayer, self).__init__() - layers = [] - layers.append( - nn.Conv2d( - in_channels, out_channels, 1, stride=1, padding=0, bias=False - ) - ) - layers.append(nn.BatchNorm2d(out_channels)) - - if nonlinear == 'relu': - layers.append(nn.ReLU(inplace=True)) - elif nonlinear == 'leakyrelu': - layers.append(nn.LeakyReLU(0.1)) - - self.layers = nn.Sequential(*layers) - - def forward(self, x): - return self.layers(x) - - -class PCB(nn.Module): - """Part-based Convolutional Baseline. - - Reference: - Sun et al. Beyond Part Models: Person Retrieval with Refined - Part Pooling (and A Strong Convolutional Baseline). ECCV 2018. - - Public keys: - - ``pcb_p4``: PCB with 4-part strips. - - ``pcb_p6``: PCB with 6-part strips. - """ - - def __init__( - self, - num_classes, - loss, - block, - layers, - parts=6, - reduced_dim=256, - nonlinear='relu', - **kwargs - ): - self.inplanes = 64 - super(PCB, self).__init__() - self.loss = loss - self.parts = parts - self.feature_dim = 512 * block.expansion - - # backbone network - self.conv1 = nn.Conv2d( - 3, 64, kernel_size=7, stride=2, padding=3, bias=False - ) - self.bn1 = nn.BatchNorm2d(64) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer(block, 64, layers[0]) - self.layer2 = self._make_layer(block, 128, layers[1], stride=2) - self.layer3 = self._make_layer(block, 256, layers[2], stride=2) - self.layer4 = self._make_layer(block, 512, layers[3], stride=1) - - # pcb layers - self.parts_avgpool = nn.AdaptiveAvgPool2d((self.parts, 1)) - self.dropout = nn.Dropout(p=0.5) - self.conv5 = DimReduceLayer( - 512 * block.expansion, reduced_dim, nonlinear=nonlinear - ) - self.feature_dim = reduced_dim - self.classifier = nn.ModuleList( - [ - nn.Linear(self.feature_dim, num_classes) - for _ in range(self.parts) - ] - ) - - self._init_params() - - def _make_layer(self, block, planes, blocks, stride=1): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d( - self.inplanes, - planes * block.expansion, - kernel_size=1, - stride=stride, - bias=False - ), - nn.BatchNorm2d(planes * block.expansion), - ) - - layers = [] - layers.append(block(self.inplanes, planes, stride, downsample)) - self.inplanes = planes * block.expansion - for i in range(1, blocks): - layers.append(block(self.inplanes, planes)) - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v_g = self.parts_avgpool(f) - - if not self.training: - v_g = F.normalize(v_g, p=2, dim=1) - return v_g.view(v_g.size(0), -1) - - v_g = self.dropout(v_g) - v_h = self.conv5(v_g) - - y = [] - for i in range(self.parts): - v_h_i = v_h[:, :, i, :] - v_h_i = v_h_i.view(v_h_i.size(0), -1) - y_i = self.classifier[i](v_h_i) - y.append(y_i) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - v_g = F.normalize(v_g, p=2, dim=1) - return y, v_g.view(v_g.size(0), -1) - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def pcb_p6(num_classes, loss='softmax', pretrained=True, **kwargs): - model = PCB( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=1, - parts=6, - reduced_dim=256, - nonlinear='relu', - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model - - -def pcb_p4(num_classes, loss='softmax', pretrained=True, **kwargs): - model = PCB( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=1, - parts=4, - reduced_dim=256, - nonlinear='relu', - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py deleted file mode 100644 index be87f3fbdf..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py +++ /dev/null @@ -1,576 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. - -Code source: https://github.com/pytorch/vision -""" -from __future__ import division, absolute_import -import torch.utils.model_zoo as model_zoo -from torch import nn - -__all__ = [ - 'resnet18', 'resnet34', 'resnet50', 'resnet101', 'resnet152', - 'resnext50_32x4d', 'resnext101_32x8d', 'resnet50_fc512' -] - -model_urls = { - 'resnet18': - 'https://download.pytorch.org/models/resnet18-5c106cde.pth', - 'resnet34': - 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', - 'resnet50': - 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': - 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': - 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', - 'resnext50_32x4d': - 'https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth', - 'resnext101_32x8d': - 'https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1, groups=1, dilation=1): - """3x3 convolution with padding""" - return nn.Conv2d( - in_planes, - out_planes, - kernel_size=3, - stride=stride, - padding=dilation, - groups=groups, - bias=False, - dilation=dilation - ) - - -def conv1x1(in_planes, out_planes, stride=1): - """1x1 convolution""" - return nn.Conv2d( - in_planes, out_planes, kernel_size=1, stride=stride, bias=False - ) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__( - self, - inplanes, - planes, - stride=1, - downsample=None, - groups=1, - base_width=64, - dilation=1, - norm_layer=None - ): - super(BasicBlock, self).__init__() - if norm_layer is None: - norm_layer = nn.BatchNorm2d - if groups != 1 or base_width != 64: - raise ValueError( - 'BasicBlock only supports groups=1 and base_width=64' - ) - if dilation > 1: - raise NotImplementedError( - "Dilation > 1 not supported in BasicBlock" - ) - # Both self.conv1 and self.downsample layers downsample the input when stride != 1 - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = norm_layer(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = norm_layer(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__( - self, - inplanes, - planes, - stride=1, - downsample=None, - groups=1, - base_width=64, - dilation=1, - norm_layer=None - ): - super(Bottleneck, self).__init__() - if norm_layer is None: - norm_layer = nn.BatchNorm2d - width = int(planes * (base_width/64.)) * groups - # Both self.conv2 and self.downsample layers downsample the input when stride != 1 - self.conv1 = conv1x1(inplanes, width) - self.bn1 = norm_layer(width) - self.conv2 = conv3x3(width, width, stride, groups, dilation) - self.bn2 = norm_layer(width) - self.conv3 = conv1x1(width, planes * self.expansion) - self.bn3 = norm_layer(planes * self.expansion) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class ResNet(nn.Module): - """Residual network. - - Reference: - - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. - - Xie et al. Aggregated Residual Transformations for Deep Neural Networks. CVPR 2017. - - Public keys: - - ``resnet18``: ResNet18. - - ``resnet34``: ResNet34. - - ``resnet50``: ResNet50. - - ``resnet101``: ResNet101. - - ``resnet152``: ResNet152. - - ``resnext50_32x4d``: ResNeXt50. - - ``resnext101_32x8d``: ResNeXt101. - - ``resnet50_fc512``: ResNet50 + FC. - """ - - def __init__( - self, - num_classes, - loss, - block, - layers, - zero_init_residual=False, - groups=1, - width_per_group=64, - replace_stride_with_dilation=None, - norm_layer=None, - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ): - super(ResNet, self).__init__() - if norm_layer is None: - norm_layer = nn.BatchNorm2d - self._norm_layer = norm_layer - self.loss = loss - self.feature_dim = 512 * block.expansion - self.inplanes = 64 - self.dilation = 1 - if replace_stride_with_dilation is None: - # each element in the tuple indicates if we should replace - # the 2x2 stride with a dilated convolution instead - replace_stride_with_dilation = [False, False, False] - if len(replace_stride_with_dilation) != 3: - raise ValueError( - "replace_stride_with_dilation should be None " - "or a 3-element tuple, got {}". - format(replace_stride_with_dilation) - ) - self.groups = groups - self.base_width = width_per_group - self.conv1 = nn.Conv2d( - 3, self.inplanes, kernel_size=7, stride=2, padding=3, bias=False - ) - self.bn1 = norm_layer(self.inplanes) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer(block, 64, layers[0]) - self.layer2 = self._make_layer( - block, - 128, - layers[1], - stride=2, - dilate=replace_stride_with_dilation[0] - ) - self.layer3 = self._make_layer( - block, - 256, - layers[2], - stride=2, - dilate=replace_stride_with_dilation[1] - ) - self.layer4 = self._make_layer( - block, - 512, - layers[3], - stride=last_stride, - dilate=replace_stride_with_dilation[2] - ) - self.global_avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = self._construct_fc_layer( - fc_dims, 512 * block.expansion, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - # Zero-initialize the last BN in each residual branch, - # so that the residual branch starts with zeros, and each residual block behaves like an identity. - # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677 - if zero_init_residual: - for m in self.modules(): - if isinstance(m, Bottleneck): - nn.init.constant_(m.bn3.weight, 0) - elif isinstance(m, BasicBlock): - nn.init.constant_(m.bn2.weight, 0) - - def _make_layer(self, block, planes, blocks, stride=1, dilate=False): - norm_layer = self._norm_layer - downsample = None - previous_dilation = self.dilation - if dilate: - self.dilation *= stride - stride = 1 - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - conv1x1(self.inplanes, planes * block.expansion, stride), - norm_layer(planes * block.expansion), - ) - - layers = [] - layers.append( - block( - self.inplanes, planes, stride, downsample, self.groups, - self.base_width, previous_dilation, norm_layer - ) - ) - self.inplanes = planes * block.expansion - for _ in range(1, blocks): - layers.append( - block( - self.inplanes, - planes, - groups=self.groups, - base_width=self.base_width, - dilation=self.dilation, - norm_layer=norm_layer - ) - ) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -"""ResNet""" - - -def resnet18(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=BasicBlock, - layers=[2, 2, 2, 2], - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet18']) - return model - - -def resnet34(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=BasicBlock, - layers=[3, 4, 6, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet34']) - return model - - -def resnet50(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model - - -def resnet101(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 23, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet101']) - return model - - -def resnet152(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 8, 36, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet152']) - return model - - -"""ResNeXt""" - - -def resnext50_32x4d(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - groups=32, - width_per_group=4, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnext50_32x4d']) - return model - - -def resnext101_32x8d(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 23, 3], - last_stride=2, - fc_dims=None, - dropout_p=None, - groups=32, - width_per_group=8, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnext101_32x8d']) - return model - - -""" -ResNet + FC -""" - - -def resnet50_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNet( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=1, - fc_dims=[512], - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py deleted file mode 100644 index 579f5ed22b..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py +++ /dev/null @@ -1,334 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Credit to https://github.com/XingangPan/IBN-Net. -""" -from __future__ import division, absolute_import -import math -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['resnet50_ibn_a'] - -model_urls = { - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1): - "3x3 convolution with padding" - return nn.Conv2d( - in_planes, - out_planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(BasicBlock, self).__init__() - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = nn.BatchNorm2d(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = nn.BatchNorm2d(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class IBN(nn.Module): - - def __init__(self, planes): - super(IBN, self).__init__() - half1 = int(planes / 2) - self.half = half1 - half2 = planes - half1 - self.IN = nn.InstanceNorm2d(half1, affine=True) - self.BN = nn.BatchNorm2d(half2) - - def forward(self, x): - split = torch.split(x, self.half, 1) - out1 = self.IN(split[0].contiguous()) - out2 = self.BN(split[1].contiguous()) - out = torch.cat((out1, out2), 1) - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, ibn=False, stride=1, downsample=None): - super(Bottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) - if ibn: - self.bn1 = IBN(planes) - else: - self.bn1 = nn.BatchNorm2d(planes) - self.conv2 = nn.Conv2d( - planes, - planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes) - self.conv3 = nn.Conv2d( - planes, planes * self.expansion, kernel_size=1, bias=False - ) - self.bn3 = nn.BatchNorm2d(planes * self.expansion) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class ResNet(nn.Module): - """Residual network + IBN layer. - - Reference: - - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. - - Pan et al. Two at Once: Enhancing Learning and Generalization - Capacities via IBN-Net. ECCV 2018. - """ - - def __init__( - self, - block, - layers, - num_classes=1000, - loss='softmax', - fc_dims=None, - dropout_p=None, - **kwargs - ): - scale = 64 - self.inplanes = scale - super(ResNet, self).__init__() - self.loss = loss - self.feature_dim = scale * 8 * block.expansion - - self.conv1 = nn.Conv2d( - 3, scale, kernel_size=7, stride=2, padding=3, bias=False - ) - self.bn1 = nn.BatchNorm2d(scale) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer(block, scale, layers[0]) - self.layer2 = self._make_layer(block, scale * 2, layers[1], stride=2) - self.layer3 = self._make_layer(block, scale * 4, layers[2], stride=2) - self.layer4 = self._make_layer(block, scale * 8, layers[3], stride=2) - self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = self._construct_fc_layer( - fc_dims, scale * 8 * block.expansion, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels - m.weight.data.normal_(0, math.sqrt(2. / n)) - elif isinstance(m, nn.BatchNorm2d): - m.weight.data.fill_(1) - m.bias.data.zero_() - elif isinstance(m, nn.InstanceNorm2d): - m.weight.data.fill_(1) - m.bias.data.zero_() - - def _make_layer(self, block, planes, blocks, stride=1): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d( - self.inplanes, - planes * block.expansion, - kernel_size=1, - stride=stride, - bias=False - ), - nn.BatchNorm2d(planes * block.expansion), - ) - - layers = [] - ibn = True - if planes == 512: - ibn = False - layers.append(block(self.inplanes, planes, ibn, stride, downsample)) - self.inplanes = planes * block.expansion - for i in range(1, blocks): - layers.append(block(self.inplanes, planes, ibn)) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.avgpool(f) - v = v.view(v.size(0), -1) - if self.fc is not None: - v = self.fc(v) - if not self.training: - return v - y = self.classifier(v) - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def resnet50_ibn_a(num_classes, loss='softmax', pretrained=False, **kwargs): - model = ResNet( - Bottleneck, [3, 4, 6, 3], num_classes=num_classes, loss=loss, **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py deleted file mode 100644 index 7026fb9e3d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py +++ /dev/null @@ -1,319 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Credit to https://github.com/XingangPan/IBN-Net. -""" -from __future__ import division, absolute_import -import math -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['resnet50_ibn_b'] - -model_urls = { - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1): - "3x3 convolution with padding" - return nn.Conv2d( - in_planes, - out_planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(BasicBlock, self).__init__() - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = nn.BatchNorm2d(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = nn.BatchNorm2d(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None, IN=False): - super(Bottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) - self.bn1 = nn.BatchNorm2d(planes) - self.conv2 = nn.Conv2d( - planes, - planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes) - self.conv3 = nn.Conv2d( - planes, planes * self.expansion, kernel_size=1, bias=False - ) - self.bn3 = nn.BatchNorm2d(planes * self.expansion) - self.IN = None - if IN: - self.IN = nn.InstanceNorm2d(planes * 4, affine=True) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - if self.IN is not None: - out = self.IN(out) - out = self.relu(out) - - return out - - -class ResNet(nn.Module): - """Residual network + IBN layer. - - Reference: - - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. - - Pan et al. Two at Once: Enhancing Learning and Generalization - Capacities via IBN-Net. ECCV 2018. - """ - - def __init__( - self, - block, - layers, - num_classes=1000, - loss='softmax', - fc_dims=None, - dropout_p=None, - **kwargs - ): - scale = 64 - self.inplanes = scale - super(ResNet, self).__init__() - self.loss = loss - self.feature_dim = scale * 8 * block.expansion - - self.conv1 = nn.Conv2d( - 3, scale, kernel_size=7, stride=2, padding=3, bias=False - ) - self.bn1 = nn.InstanceNorm2d(scale, affine=True) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer( - block, scale, layers[0], stride=1, IN=True - ) - self.layer2 = self._make_layer( - block, scale * 2, layers[1], stride=2, IN=True - ) - self.layer3 = self._make_layer(block, scale * 4, layers[2], stride=2) - self.layer4 = self._make_layer(block, scale * 8, layers[3], stride=2) - self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = self._construct_fc_layer( - fc_dims, scale * 8 * block.expansion, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels - m.weight.data.normal_(0, math.sqrt(2. / n)) - elif isinstance(m, nn.BatchNorm2d): - m.weight.data.fill_(1) - m.bias.data.zero_() - elif isinstance(m, nn.InstanceNorm2d): - m.weight.data.fill_(1) - m.bias.data.zero_() - - def _make_layer(self, block, planes, blocks, stride=1, IN=False): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d( - self.inplanes, - planes * block.expansion, - kernel_size=1, - stride=stride, - bias=False - ), - nn.BatchNorm2d(planes * block.expansion), - ) - - layers = [] - layers.append(block(self.inplanes, planes, stride, downsample)) - self.inplanes = planes * block.expansion - for i in range(1, blocks - 1): - layers.append(block(self.inplanes, planes)) - layers.append(block(self.inplanes, planes, IN=IN)) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.avgpool(f) - v = v.view(v.size(0), -1) - if self.fc is not None: - v = self.fc(v) - if not self.training: - return v - y = self.classifier(v) - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def resnet50_ibn_b(num_classes, loss='softmax', pretrained=False, **kwargs): - model = ResNet( - Bottleneck, [3, 4, 6, 3], num_classes=num_classes, loss=loss, **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py deleted file mode 100644 index 87130a2fb2..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py +++ /dev/null @@ -1,354 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.utils.model_zoo as model_zoo -from torch import nn - -__all__ = ['resnet50mid'] - -model_urls = { - 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', - 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1): - """3x3 convolution with padding""" - return nn.Conv2d( - in_planes, - out_planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(BasicBlock, self).__init__() - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = nn.BatchNorm2d(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = nn.BatchNorm2d(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None): - super(Bottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) - self.bn1 = nn.BatchNorm2d(planes) - self.conv2 = nn.Conv2d( - planes, - planes, - kernel_size=3, - stride=stride, - padding=1, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes) - self.conv3 = nn.Conv2d( - planes, planes * self.expansion, kernel_size=1, bias=False - ) - self.bn3 = nn.BatchNorm2d(planes * self.expansion) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out += residual - out = self.relu(out) - - return out - - -class ResNetMid(nn.Module): - """Residual network + mid-level features. - - Reference: - Yu et al. The Devil is in the Middle: Exploiting Mid-level Representations for - Cross-Domain Instance Matching. arXiv:1711.08106. - - Public keys: - - ``resnet50mid``: ResNet50 + mid-level feature fusion. - """ - - def __init__( - self, - num_classes, - loss, - block, - layers, - last_stride=2, - fc_dims=None, - **kwargs - ): - self.inplanes = 64 - super(ResNetMid, self).__init__() - self.loss = loss - self.feature_dim = 512 * block.expansion - - # backbone network - self.conv1 = nn.Conv2d( - 3, 64, kernel_size=7, stride=2, padding=3, bias=False - ) - self.bn1 = nn.BatchNorm2d(64) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer(block, 64, layers[0]) - self.layer2 = self._make_layer(block, 128, layers[1], stride=2) - self.layer3 = self._make_layer(block, 256, layers[2], stride=2) - self.layer4 = self._make_layer( - block, 512, layers[3], stride=last_stride - ) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - assert fc_dims is not None - self.fc_fusion = self._construct_fc_layer( - fc_dims, 512 * block.expansion * 2 - ) - self.feature_dim += 512 * block.expansion - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _make_layer(self, block, planes, blocks, stride=1): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d( - self.inplanes, - planes * block.expansion, - kernel_size=1, - stride=stride, - bias=False - ), - nn.BatchNorm2d(planes * block.expansion), - ) - - layers = [] - layers.append(block(self.inplanes, planes, stride, downsample)) - self.inplanes = planes * block.expansion - for i in range(1, blocks): - layers.append(block(self.inplanes, planes)) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x4a = self.layer4[0](x) - x4b = self.layer4[1](x4a) - x4c = self.layer4[2](x4b) - return x4a, x4b, x4c - - def forward(self, x): - x4a, x4b, x4c = self.featuremaps(x) - - v4a = self.global_avgpool(x4a) - v4b = self.global_avgpool(x4b) - v4c = self.global_avgpool(x4c) - v4ab = torch.cat([v4a, v4b], 1) - v4ab = v4ab.view(v4ab.size(0), -1) - v4ab = self.fc_fusion(v4ab) - v4c = v4c.view(v4c.size(0), -1) - v = torch.cat([v4ab, v4c], 1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -""" -Residual network configurations: --- -resnet18: block=BasicBlock, layers=[2, 2, 2, 2] -resnet34: block=BasicBlock, layers=[3, 4, 6, 3] -resnet50: block=Bottleneck, layers=[3, 4, 6, 3] -resnet101: block=Bottleneck, layers=[3, 4, 23, 3] -resnet152: block=Bottleneck, layers=[3, 8, 36, 3] -""" - - -def resnet50mid(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ResNetMid( - num_classes=num_classes, - loss=loss, - block=Bottleneck, - layers=[3, 4, 6, 3], - last_stride=2, - fc_dims=[1024], - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['resnet50']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py deleted file mode 100644 index ef38f0e56f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py +++ /dev/null @@ -1,735 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import math -from collections import OrderedDict -import torch.nn as nn -from torch.utils import model_zoo - -__all__ = [ - 'senet154', 'se_resnet50', 'se_resnet101', 'se_resnet152', - 'se_resnext50_32x4d', 'se_resnext101_32x4d', 'se_resnet50_fc512' -] -""" -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" - -pretrained_settings = { - 'senet154': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/senet154-c7b49a05.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, - 'se_resnet50': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet50-ce0d4300.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, - 'se_resnet101': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet101-7e38fcc6.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, - 'se_resnet152': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet152-d17c99b7.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, - 'se_resnext50_32x4d': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnext50_32x4d-a260b3a4.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, - 'se_resnext101_32x4d': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnext101_32x4d-3b2fe3d8.pth', - 'input_space': 'RGB', - 'input_size': [3, 224, 224], - 'input_range': [0, 1], - 'mean': [0.485, 0.456, 0.406], - 'std': [0.229, 0.224, 0.225], - 'num_classes': 1000 - } - }, -} - - -class SEModule(nn.Module): - - def __init__(self, channels, reduction): - super(SEModule, self).__init__() - self.avg_pool = nn.AdaptiveAvgPool2d(1) - self.fc1 = nn.Conv2d( - channels, channels // reduction, kernel_size=1, padding=0 - ) - self.relu = nn.ReLU(inplace=True) - self.fc2 = nn.Conv2d( - channels // reduction, channels, kernel_size=1, padding=0 - ) - self.sigmoid = nn.Sigmoid() - - def forward(self, x): - module_input = x - x = self.avg_pool(x) - x = self.fc1(x) - x = self.relu(x) - x = self.fc2(x) - x = self.sigmoid(x) - return module_input * x - - -class Bottleneck(nn.Module): - """ - Base class for bottlenecks that implements `forward()` method. - """ - - def forward(self, x): - residual = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - residual = self.downsample(x) - - out = self.se_module(out) + residual - out = self.relu(out) - - return out - - -class SEBottleneck(Bottleneck): - """ - Bottleneck for SENet154. - """ - expansion = 4 - - def __init__( - self, inplanes, planes, groups, reduction, stride=1, downsample=None - ): - super(SEBottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes * 2, kernel_size=1, bias=False) - self.bn1 = nn.BatchNorm2d(planes * 2) - self.conv2 = nn.Conv2d( - planes * 2, - planes * 4, - kernel_size=3, - stride=stride, - padding=1, - groups=groups, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes * 4) - self.conv3 = nn.Conv2d( - planes * 4, planes * 4, kernel_size=1, bias=False - ) - self.bn3 = nn.BatchNorm2d(planes * 4) - self.relu = nn.ReLU(inplace=True) - self.se_module = SEModule(planes * 4, reduction=reduction) - self.downsample = downsample - self.stride = stride - - -class SEResNetBottleneck(Bottleneck): - """ - ResNet bottleneck with a Squeeze-and-Excitation module. It follows Caffe - implementation and uses `stride=stride` in `conv1` and not in `conv2` - (the latter is used in the torchvision implementation of ResNet). - """ - expansion = 4 - - def __init__( - self, inplanes, planes, groups, reduction, stride=1, downsample=None - ): - super(SEResNetBottleneck, self).__init__() - self.conv1 = nn.Conv2d( - inplanes, planes, kernel_size=1, bias=False, stride=stride - ) - self.bn1 = nn.BatchNorm2d(planes) - self.conv2 = nn.Conv2d( - planes, - planes, - kernel_size=3, - padding=1, - groups=groups, - bias=False - ) - self.bn2 = nn.BatchNorm2d(planes) - self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False) - self.bn3 = nn.BatchNorm2d(planes * 4) - self.relu = nn.ReLU(inplace=True) - self.se_module = SEModule(planes * 4, reduction=reduction) - self.downsample = downsample - self.stride = stride - - -class SEResNeXtBottleneck(Bottleneck): - """ResNeXt bottleneck type C with a Squeeze-and-Excitation module""" - expansion = 4 - - def __init__( - self, - inplanes, - planes, - groups, - reduction, - stride=1, - downsample=None, - base_width=4 - ): - super(SEResNeXtBottleneck, self).__init__() - width = int(math.floor(planes * (base_width/64.)) * groups) - self.conv1 = nn.Conv2d( - inplanes, width, kernel_size=1, bias=False, stride=1 - ) - self.bn1 = nn.BatchNorm2d(width) - self.conv2 = nn.Conv2d( - width, - width, - kernel_size=3, - stride=stride, - padding=1, - groups=groups, - bias=False - ) - self.bn2 = nn.BatchNorm2d(width) - self.conv3 = nn.Conv2d(width, planes * 4, kernel_size=1, bias=False) - self.bn3 = nn.BatchNorm2d(planes * 4) - self.relu = nn.ReLU(inplace=True) - self.se_module = SEModule(planes * 4, reduction=reduction) - self.downsample = downsample - self.stride = stride - - -class SENet(nn.Module): - """Squeeze-and-excitation network. - - Reference: - Hu et al. Squeeze-and-Excitation Networks. CVPR 2018. - - Public keys: - - ``senet154``: SENet154. - - ``se_resnet50``: ResNet50 + SE. - - ``se_resnet101``: ResNet101 + SE. - - ``se_resnet152``: ResNet152 + SE. - - ``se_resnext50_32x4d``: ResNeXt50 (groups=32, width=4) + SE. - - ``se_resnext101_32x4d``: ResNeXt101 (groups=32, width=4) + SE. - - ``se_resnet50_fc512``: (ResNet50 + SE) + FC. - """ - - def __init__( - self, - num_classes, - loss, - block, - layers, - groups, - reduction, - dropout_p=0.2, - inplanes=128, - input_3x3=True, - downsample_kernel_size=3, - downsample_padding=1, - last_stride=2, - fc_dims=None, - **kwargs - ): - """ - Parameters - ---------- - block (nn.Module): Bottleneck class. - - For SENet154: SEBottleneck - - For SE-ResNet models: SEResNetBottleneck - - For SE-ResNeXt models: SEResNeXtBottleneck - layers (list of ints): Number of residual blocks for 4 layers of the - network (layer1...layer4). - groups (int): Number of groups for the 3x3 convolution in each - bottleneck block. - - For SENet154: 64 - - For SE-ResNet models: 1 - - For SE-ResNeXt models: 32 - reduction (int): Reduction ratio for Squeeze-and-Excitation modules. - - For all models: 16 - dropout_p (float or None): Drop probability for the Dropout layer. - If `None` the Dropout layer is not used. - - For SENet154: 0.2 - - For SE-ResNet models: None - - For SE-ResNeXt models: None - inplanes (int): Number of input channels for layer1. - - For SENet154: 128 - - For SE-ResNet models: 64 - - For SE-ResNeXt models: 64 - input_3x3 (bool): If `True`, use three 3x3 convolutions instead of - a single 7x7 convolution in layer0. - - For SENet154: True - - For SE-ResNet models: False - - For SE-ResNeXt models: False - downsample_kernel_size (int): Kernel size for downsampling convolutions - in layer2, layer3 and layer4. - - For SENet154: 3 - - For SE-ResNet models: 1 - - For SE-ResNeXt models: 1 - downsample_padding (int): Padding for downsampling convolutions in - layer2, layer3 and layer4. - - For SENet154: 1 - - For SE-ResNet models: 0 - - For SE-ResNeXt models: 0 - num_classes (int): Number of outputs in `classifier` layer. - """ - super(SENet, self).__init__() - self.inplanes = inplanes - self.loss = loss - - if input_3x3: - layer0_modules = [ - ( - 'conv1', - nn.Conv2d(3, 64, 3, stride=2, padding=1, bias=False) - ), - ('bn1', nn.BatchNorm2d(64)), - ('relu1', nn.ReLU(inplace=True)), - ( - 'conv2', - nn.Conv2d(64, 64, 3, stride=1, padding=1, bias=False) - ), - ('bn2', nn.BatchNorm2d(64)), - ('relu2', nn.ReLU(inplace=True)), - ( - 'conv3', - nn.Conv2d( - 64, inplanes, 3, stride=1, padding=1, bias=False - ) - ), - ('bn3', nn.BatchNorm2d(inplanes)), - ('relu3', nn.ReLU(inplace=True)), - ] - else: - layer0_modules = [ - ( - 'conv1', - nn.Conv2d( - 3, - inplanes, - kernel_size=7, - stride=2, - padding=3, - bias=False - ) - ), - ('bn1', nn.BatchNorm2d(inplanes)), - ('relu1', nn.ReLU(inplace=True)), - ] - # To preserve compatibility with Caffe weights `ceil_mode=True` - # is used instead of `padding=1`. - layer0_modules.append( - ('pool', nn.MaxPool2d(3, stride=2, ceil_mode=True)) - ) - self.layer0 = nn.Sequential(OrderedDict(layer0_modules)) - self.layer1 = self._make_layer( - block, - planes=64, - blocks=layers[0], - groups=groups, - reduction=reduction, - downsample_kernel_size=1, - downsample_padding=0 - ) - self.layer2 = self._make_layer( - block, - planes=128, - blocks=layers[1], - stride=2, - groups=groups, - reduction=reduction, - downsample_kernel_size=downsample_kernel_size, - downsample_padding=downsample_padding - ) - self.layer3 = self._make_layer( - block, - planes=256, - blocks=layers[2], - stride=2, - groups=groups, - reduction=reduction, - downsample_kernel_size=downsample_kernel_size, - downsample_padding=downsample_padding - ) - self.layer4 = self._make_layer( - block, - planes=512, - blocks=layers[3], - stride=last_stride, - groups=groups, - reduction=reduction, - downsample_kernel_size=downsample_kernel_size, - downsample_padding=downsample_padding - ) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = self._construct_fc_layer( - fc_dims, 512 * block.expansion, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - def _make_layer( - self, - block, - planes, - blocks, - groups, - reduction, - stride=1, - downsample_kernel_size=1, - downsample_padding=0 - ): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d( - self.inplanes, - planes * block.expansion, - kernel_size=downsample_kernel_size, - stride=stride, - padding=downsample_padding, - bias=False - ), - nn.BatchNorm2d(planes * block.expansion), - ) - - layers = [] - layers.append( - block( - self.inplanes, planes, groups, reduction, stride, downsample - ) - ) - self.inplanes = planes * block.expansion - for i in range(1, blocks): - layers.append(block(self.inplanes, planes, groups, reduction)) - - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """ - Construct fully connected layer - - - fc_dims (list or tuple): dimensions of fc layers, if None, - no fc layers are constructed - - input_dim (int): input dimension - - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def featuremaps(self, x): - x = self.layer0(x) - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def senet154(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEBottleneck, - layers=[3, 8, 36, 3], - groups=64, - reduction=16, - dropout_p=0.2, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['senet154']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model - - -def se_resnet50(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNetBottleneck, - layers=[3, 4, 6, 3], - groups=1, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnet50']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model - - -def se_resnet50_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNetBottleneck, - layers=[3, 4, 6, 3], - groups=1, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=1, - fc_dims=[512], - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnet50']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model - - -def se_resnet101(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNetBottleneck, - layers=[3, 4, 23, 3], - groups=1, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnet101']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model - - -def se_resnet152(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNetBottleneck, - layers=[3, 8, 36, 3], - groups=1, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnet152']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model - - -def se_resnext50_32x4d(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNeXtBottleneck, - layers=[3, 4, 6, 3], - groups=32, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnext50_32x4d']['imagenet']['url' - ] - init_pretrained_weights(model, model_url) - return model - - -def se_resnext101_32x4d( - num_classes, loss='softmax', pretrained=True, **kwargs -): - model = SENet( - num_classes=num_classes, - loss=loss, - block=SEResNeXtBottleneck, - layers=[3, 4, 23, 3], - groups=32, - reduction=16, - dropout_p=None, - inplanes=64, - input_3x3=False, - downsample_kernel_size=1, - downsample_padding=0, - last_stride=2, - fc_dims=None, - **kwargs - ) - if pretrained: - model_url = pretrained_settings['se_resnext101_32x4d']['imagenet'][ - 'url'] - init_pretrained_weights(model, model_url) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py deleted file mode 100644 index 3e05ff7bc5..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py +++ /dev/null @@ -1,245 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['shufflenet'] - -model_urls = { - # training epoch = 90, top1 = 61.8 - 'imagenet': - 'https://mega.nz/#!RDpUlQCY!tr_5xBEkelzDjveIYBBcGcovNCOrgfiJO9kiidz9fZM', -} - - -class ChannelShuffle(nn.Module): - - def __init__(self, num_groups): - super(ChannelShuffle, self).__init__() - self.g = num_groups - - def forward(self, x): - b, c, h, w = x.size() - n = c // self.g - # reshape - x = x.view(b, self.g, n, h, w) - # transpose - x = x.permute(0, 2, 1, 3, 4).contiguous() - # flatten - x = x.view(b, c, h, w) - return x - - -class Bottleneck(nn.Module): - - def __init__( - self, - in_channels, - out_channels, - stride, - num_groups, - group_conv1x1=True - ): - super(Bottleneck, self).__init__() - assert stride in [1, 2], 'Warning: stride must be either 1 or 2' - self.stride = stride - mid_channels = out_channels // 4 - if stride == 2: - out_channels -= in_channels - # group conv is not applied to first conv1x1 at stage 2 - num_groups_conv1x1 = num_groups if group_conv1x1 else 1 - self.conv1 = nn.Conv2d( - in_channels, - mid_channels, - 1, - groups=num_groups_conv1x1, - bias=False - ) - self.bn1 = nn.BatchNorm2d(mid_channels) - self.shuffle1 = ChannelShuffle(num_groups) - self.conv2 = nn.Conv2d( - mid_channels, - mid_channels, - 3, - stride=stride, - padding=1, - groups=mid_channels, - bias=False - ) - self.bn2 = nn.BatchNorm2d(mid_channels) - self.conv3 = nn.Conv2d( - mid_channels, out_channels, 1, groups=num_groups, bias=False - ) - self.bn3 = nn.BatchNorm2d(out_channels) - if stride == 2: - self.shortcut = nn.AvgPool2d(3, stride=2, padding=1) - - def forward(self, x): - out = F.relu(self.bn1(self.conv1(x))) - out = self.shuffle1(out) - out = self.bn2(self.conv2(out)) - out = self.bn3(self.conv3(out)) - if self.stride == 2: - res = self.shortcut(x) - out = F.relu(torch.cat([res, out], 1)) - else: - out = F.relu(x + out) - return out - - -# configuration of (num_groups: #out_channels) based on Table 1 in the paper -cfg = { - 1: [144, 288, 576], - 2: [200, 400, 800], - 3: [240, 480, 960], - 4: [272, 544, 1088], - 8: [384, 768, 1536], -} - - -class ShuffleNet(nn.Module): - """ShuffleNet. - - Reference: - Zhang et al. ShuffleNet: An Extremely Efficient Convolutional Neural - Network for Mobile Devices. CVPR 2018. - - Public keys: - - ``shufflenet``: ShuffleNet (groups=3). - """ - - def __init__(self, num_classes, loss='softmax', num_groups=3, **kwargs): - super(ShuffleNet, self).__init__() - self.loss = loss - - self.conv1 = nn.Sequential( - nn.Conv2d(3, 24, 3, stride=2, padding=1, bias=False), - nn.BatchNorm2d(24), - nn.ReLU(), - nn.MaxPool2d(3, stride=2, padding=1), - ) - - self.stage2 = nn.Sequential( - Bottleneck( - 24, cfg[num_groups][0], 2, num_groups, group_conv1x1=False - ), - Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), - Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), - Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), - ) - - self.stage3 = nn.Sequential( - Bottleneck(cfg[num_groups][0], cfg[num_groups][1], 2, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), - ) - - self.stage4 = nn.Sequential( - Bottleneck(cfg[num_groups][1], cfg[num_groups][2], 2, num_groups), - Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), - Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), - Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), - ) - - self.classifier = nn.Linear(cfg[num_groups][2], num_classes) - self.feat_dim = cfg[num_groups][2] - - def forward(self, x): - x = self.conv1(x) - x = self.stage2(x) - x = self.stage3(x) - x = self.stage4(x) - x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), -1) - - if not self.training: - return x - - y = self.classifier(x) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, x - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def shufflenet(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ShuffleNet(num_classes, loss, **kwargs) - if pretrained: - # init_pretrained_weights(model, model_urls['imagenet']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['imagenet']) - ) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py deleted file mode 100644 index 2b9fa4d403..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py +++ /dev/null @@ -1,307 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code source: https://github.com/pytorch/vision -""" -from __future__ import division, absolute_import -import torch -import torch.utils.model_zoo as model_zoo -from torch import nn - -__all__ = [ - 'shufflenet_v2_x0_5', 'shufflenet_v2_x1_0', 'shufflenet_v2_x1_5', - 'shufflenet_v2_x2_0' -] - -model_urls = { - 'shufflenetv2_x0.5': - 'https://download.pytorch.org/models/shufflenetv2_x0.5-f707e7126e.pth', - 'shufflenetv2_x1.0': - 'https://download.pytorch.org/models/shufflenetv2_x1-5666bf0f80.pth', - 'shufflenetv2_x1.5': None, - 'shufflenetv2_x2.0': None, -} - - -def channel_shuffle(x, groups): - batchsize, num_channels, height, width = x.data.size() - channels_per_group = num_channels // groups - - # reshape - x = x.view(batchsize, groups, channels_per_group, height, width) - - x = torch.transpose(x, 1, 2).contiguous() - - # flatten - x = x.view(batchsize, -1, height, width) - - return x - - -class InvertedResidual(nn.Module): - - def __init__(self, inp, oup, stride): - super(InvertedResidual, self).__init__() - - if not (1 <= stride <= 3): - raise ValueError('illegal stride value') - self.stride = stride - - branch_features = oup // 2 - assert (self.stride != 1) or (inp == branch_features << 1) - - if self.stride > 1: - self.branch1 = nn.Sequential( - self.depthwise_conv( - inp, inp, kernel_size=3, stride=self.stride, padding=1 - ), - nn.BatchNorm2d(inp), - nn.Conv2d( - inp, - branch_features, - kernel_size=1, - stride=1, - padding=0, - bias=False - ), - nn.BatchNorm2d(branch_features), - nn.ReLU(inplace=True), - ) - - self.branch2 = nn.Sequential( - nn.Conv2d( - inp if (self.stride > 1) else branch_features, - branch_features, - kernel_size=1, - stride=1, - padding=0, - bias=False - ), - nn.BatchNorm2d(branch_features), - nn.ReLU(inplace=True), - self.depthwise_conv( - branch_features, - branch_features, - kernel_size=3, - stride=self.stride, - padding=1 - ), - nn.BatchNorm2d(branch_features), - nn.Conv2d( - branch_features, - branch_features, - kernel_size=1, - stride=1, - padding=0, - bias=False - ), - nn.BatchNorm2d(branch_features), - nn.ReLU(inplace=True), - ) - - @staticmethod - def depthwise_conv(i, o, kernel_size, stride=1, padding=0, bias=False): - return nn.Conv2d( - i, o, kernel_size, stride, padding, bias=bias, groups=i - ) - - def forward(self, x): - if self.stride == 1: - x1, x2 = x.chunk(2, dim=1) - out = torch.cat((x1, self.branch2(x2)), dim=1) - else: - out = torch.cat((self.branch1(x), self.branch2(x)), dim=1) - - out = channel_shuffle(out, 2) - - return out - - -class ShuffleNetV2(nn.Module): - """ShuffleNetV2. - - Reference: - Ma et al. ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. ECCV 2018. - - Public keys: - - ``shufflenet_v2_x0_5``: ShuffleNetV2 x0.5. - - ``shufflenet_v2_x1_0``: ShuffleNetV2 x1.0. - - ``shufflenet_v2_x1_5``: ShuffleNetV2 x1.5. - - ``shufflenet_v2_x2_0``: ShuffleNetV2 x2.0. - """ - - def __init__( - self, num_classes, loss, stages_repeats, stages_out_channels, **kwargs - ): - super(ShuffleNetV2, self).__init__() - self.loss = loss - - if len(stages_repeats) != 3: - raise ValueError( - 'expected stages_repeats as list of 3 positive ints' - ) - if len(stages_out_channels) != 5: - raise ValueError( - 'expected stages_out_channels as list of 5 positive ints' - ) - self._stage_out_channels = stages_out_channels - - input_channels = 3 - output_channels = self._stage_out_channels[0] - self.conv1 = nn.Sequential( - nn.Conv2d(input_channels, output_channels, 3, 2, 1, bias=False), - nn.BatchNorm2d(output_channels), - nn.ReLU(inplace=True), - ) - input_channels = output_channels - - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - - stage_names = ['stage{}'.format(i) for i in [2, 3, 4]] - for name, repeats, output_channels in zip( - stage_names, stages_repeats, self._stage_out_channels[1:] - ): - seq = [InvertedResidual(input_channels, output_channels, 2)] - for i in range(repeats - 1): - seq.append( - InvertedResidual(output_channels, output_channels, 1) - ) - setattr(self, name, nn.Sequential(*seq)) - input_channels = output_channels - - output_channels = self._stage_out_channels[-1] - self.conv5 = nn.Sequential( - nn.Conv2d(input_channels, output_channels, 1, 1, 0, bias=False), - nn.BatchNorm2d(output_channels), - nn.ReLU(inplace=True), - ) - self.global_avgpool = nn.AdaptiveAvgPool2d((1, 1)) - - self.classifier = nn.Linear(output_channels, num_classes) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.maxpool(x) - x = self.stage2(x) - x = self.stage3(x) - x = self.stage4(x) - x = self.conv5(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - if model_url is None: - import warnings - warnings.warn( - 'ImageNet pretrained weights are unavailable for this model' - ) - return - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def shufflenet_v2_x0_5(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ShuffleNetV2( - num_classes, loss, [4, 8, 4], [24, 48, 96, 192, 1024], **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['shufflenetv2_x0.5']) - return model - - -def shufflenet_v2_x1_0(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ShuffleNetV2( - num_classes, loss, [4, 8, 4], [24, 116, 232, 464, 1024], **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['shufflenetv2_x1.0']) - return model - - -def shufflenet_v2_x1_5(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ShuffleNetV2( - num_classes, loss, [4, 8, 4], [24, 176, 352, 704, 1024], **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['shufflenetv2_x1.5']) - return model - - -def shufflenet_v2_x2_0(num_classes, loss='softmax', pretrained=True, **kwargs): - model = ShuffleNetV2( - num_classes, loss, [4, 8, 4], [24, 244, 488, 976, 2048], **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['shufflenetv2_x2.0']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py deleted file mode 100644 index 2b9ebb56f2..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py +++ /dev/null @@ -1,281 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code source: https://github.com/pytorch/vision -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['squeezenet1_0', 'squeezenet1_1', 'squeezenet1_0_fc512'] - -model_urls = { - 'squeezenet1_0': - 'https://download.pytorch.org/models/squeezenet1_0-a815701f.pth', - 'squeezenet1_1': - 'https://download.pytorch.org/models/squeezenet1_1-f364aa15.pth', -} - - -class Fire(nn.Module): - - def __init__( - self, inplanes, squeeze_planes, expand1x1_planes, expand3x3_planes - ): - super(Fire, self).__init__() - self.inplanes = inplanes - self.squeeze = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1) - self.squeeze_activation = nn.ReLU(inplace=True) - self.expand1x1 = nn.Conv2d( - squeeze_planes, expand1x1_planes, kernel_size=1 - ) - self.expand1x1_activation = nn.ReLU(inplace=True) - self.expand3x3 = nn.Conv2d( - squeeze_planes, expand3x3_planes, kernel_size=3, padding=1 - ) - self.expand3x3_activation = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.squeeze_activation(self.squeeze(x)) - return torch.cat( - [ - self.expand1x1_activation(self.expand1x1(x)), - self.expand3x3_activation(self.expand3x3(x)) - ], 1 - ) - - -class SqueezeNet(nn.Module): - """SqueezeNet. - - Reference: - Iandola et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters - and< 0.5 MB model size. arXiv:1602.07360. - - Public keys: - - ``squeezenet1_0``: SqueezeNet (version=1.0). - - ``squeezenet1_1``: SqueezeNet (version=1.1). - - ``squeezenet1_0_fc512``: SqueezeNet (version=1.0) + FC. - """ - - def __init__( - self, - num_classes, - loss, - version=1.0, - fc_dims=None, - dropout_p=None, - **kwargs - ): - super(SqueezeNet, self).__init__() - self.loss = loss - self.feature_dim = 512 - - if version not in [1.0, 1.1]: - raise ValueError( - 'Unsupported SqueezeNet version {version}:' - '1.0 or 1.1 expected'.format(version=version) - ) - - if version == 1.0: - self.features = nn.Sequential( - nn.Conv2d(3, 96, kernel_size=7, stride=2), - nn.ReLU(inplace=True), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(96, 16, 64, 64), - Fire(128, 16, 64, 64), - Fire(128, 32, 128, 128), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(256, 32, 128, 128), - Fire(256, 48, 192, 192), - Fire(384, 48, 192, 192), - Fire(384, 64, 256, 256), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(512, 64, 256, 256), - ) - else: - self.features = nn.Sequential( - nn.Conv2d(3, 64, kernel_size=3, stride=2), - nn.ReLU(inplace=True), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(64, 16, 64, 64), - Fire(128, 16, 64, 64), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(128, 32, 128, 128), - Fire(256, 32, 128, 128), - nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), - Fire(256, 48, 192, 192), - Fire(384, 48, 192, 192), - Fire(384, 64, 256, 256), - Fire(512, 64, 256, 256), - ) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = self._construct_fc_layer(fc_dims, 512, dropout_p) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x): - f = self.features(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url, map_location=None) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def squeezenet1_0(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SqueezeNet( - num_classes, loss, version=1.0, fc_dims=None, dropout_p=None, **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['squeezenet1_0']) - return model - - -def squeezenet1_0_fc512( - num_classes, loss='softmax', pretrained=True, **kwargs -): - model = SqueezeNet( - num_classes, - loss, - version=1.0, - fc_dims=[512], - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['squeezenet1_0']) - return model - - -def squeezenet1_1(num_classes, loss='softmax', pretrained=True, **kwargs): - model = SqueezeNet( - num_classes, loss, version=1.1, fc_dims=None, dropout_p=None, **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['squeezenet1_1']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py deleted file mode 100644 index ed7039facb..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py +++ /dev/null @@ -1,391 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch.nn as nn -import torch.nn.functional as F -import torch.utils.model_zoo as model_zoo - -__all__ = ['xception'] - -pretrained_settings = { - 'xception': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/xception-43020ad28.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000, - 'scale': - 0.8975 # The resize parameter of the validation transform should be 333, and make sure to center crop at 299x299 - } - } -} - - -class SeparableConv2d(nn.Module): - - def __init__( - self, - in_channels, - out_channels, - kernel_size=1, - stride=1, - padding=0, - dilation=1, - bias=False - ): - super(SeparableConv2d, self).__init__() - - self.conv1 = nn.Conv2d( - in_channels, - in_channels, - kernel_size, - stride, - padding, - dilation, - groups=in_channels, - bias=bias - ) - self.pointwise = nn.Conv2d( - in_channels, out_channels, 1, 1, 0, 1, 1, bias=bias - ) - - def forward(self, x): - x = self.conv1(x) - x = self.pointwise(x) - return x - - -class Block(nn.Module): - - def __init__( - self, - in_filters, - out_filters, - reps, - strides=1, - start_with_relu=True, - grow_first=True - ): - super(Block, self).__init__() - - if out_filters != in_filters or strides != 1: - self.skip = nn.Conv2d( - in_filters, out_filters, 1, stride=strides, bias=False - ) - self.skipbn = nn.BatchNorm2d(out_filters) - else: - self.skip = None - - self.relu = nn.ReLU(inplace=True) - rep = [] - - filters = in_filters - if grow_first: - rep.append(self.relu) - rep.append( - SeparableConv2d( - in_filters, - out_filters, - 3, - stride=1, - padding=1, - bias=False - ) - ) - rep.append(nn.BatchNorm2d(out_filters)) - filters = out_filters - - for i in range(reps - 1): - rep.append(self.relu) - rep.append( - SeparableConv2d( - filters, filters, 3, stride=1, padding=1, bias=False - ) - ) - rep.append(nn.BatchNorm2d(filters)) - - if not grow_first: - rep.append(self.relu) - rep.append( - SeparableConv2d( - in_filters, - out_filters, - 3, - stride=1, - padding=1, - bias=False - ) - ) - rep.append(nn.BatchNorm2d(out_filters)) - - if not start_with_relu: - rep = rep[1:] - else: - rep[0] = nn.ReLU(inplace=False) - - if strides != 1: - rep.append(nn.MaxPool2d(3, strides, 1)) - self.rep = nn.Sequential(*rep) - - def forward(self, inp): - x = self.rep(inp) - - if self.skip is not None: - skip = self.skip(inp) - skip = self.skipbn(skip) - else: - skip = inp - - x += skip - return x - - -class Xception(nn.Module): - """Xception. - - Reference: - Chollet. Xception: Deep Learning with Depthwise - Separable Convolutions. CVPR 2017. - - Public keys: - - ``xception``: Xception. - """ - - def __init__( - self, num_classes, loss, fc_dims=None, dropout_p=None, **kwargs - ): - super(Xception, self).__init__() - self.loss = loss - - self.conv1 = nn.Conv2d(3, 32, 3, 2, 0, bias=False) - self.bn1 = nn.BatchNorm2d(32) - - self.conv2 = nn.Conv2d(32, 64, 3, bias=False) - self.bn2 = nn.BatchNorm2d(64) - - self.block1 = Block( - 64, 128, 2, 2, start_with_relu=False, grow_first=True - ) - self.block2 = Block( - 128, 256, 2, 2, start_with_relu=True, grow_first=True - ) - self.block3 = Block( - 256, 728, 2, 2, start_with_relu=True, grow_first=True - ) - - self.block4 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block5 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block6 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block7 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - - self.block8 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block9 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block10 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - self.block11 = Block( - 728, 728, 3, 1, start_with_relu=True, grow_first=True - ) - - self.block12 = Block( - 728, 1024, 2, 2, start_with_relu=True, grow_first=False - ) - - self.conv3 = SeparableConv2d(1024, 1536, 3, 1, 1) - self.bn3 = nn.BatchNorm2d(1536) - - self.conv4 = SeparableConv2d(1536, 2048, 3, 1, 1) - self.bn4 = nn.BatchNorm2d(2048) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.feature_dim = 2048 - self.fc = self._construct_fc_layer(fc_dims, 2048, dropout_p) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer. - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, input): - x = self.conv1(input) - x = self.bn1(x) - x = F.relu(x, inplace=True) - - x = self.conv2(x) - x = self.bn2(x) - x = F.relu(x, inplace=True) - - x = self.block1(x) - x = self.block2(x) - x = self.block3(x) - x = self.block4(x) - x = self.block5(x) - x = self.block6(x) - x = self.block7(x) - x = self.block8(x) - x = self.block9(x) - x = self.block10(x) - x = self.block11(x) - x = self.block12(x) - - x = self.conv3(x) - x = self.bn3(x) - x = F.relu(x, inplace=True) - - x = self.conv4(x) - x = self.bn4(x) - x = F.relu(x, inplace=True) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initialize models with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def xception(num_classes, loss='softmax', pretrained=True, **kwargs): - model = Xception(num_classes, loss, fc_dims=None, dropout_p=None, **kwargs) - if pretrained: - model_url = pretrained_settings['xception']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py deleted file mode 100644 index 24c0fdc8ef..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py +++ /dev/null @@ -1,51 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .optimizer import build_optimizer -from .lr_scheduler import build_lr_scheduler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py deleted file mode 100644 index da06eb5e8f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py +++ /dev/null @@ -1,115 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import -import torch - -AVAI_SCH = ['single_step', 'multi_step', 'cosine'] - - -def build_lr_scheduler( - optimizer, lr_scheduler='single_step', stepsize=1, gamma=0.1, max_epoch=1 -): - """A function wrapper for building a learning rate scheduler. - - Args: - optimizer (Optimizer): an Optimizer. - lr_scheduler (str, optional): learning rate scheduler method. Default is single_step. - stepsize (int or list, optional): step size to decay learning rate. When ``lr_scheduler`` - is "single_step", ``stepsize`` should be an integer. When ``lr_scheduler`` is - "multi_step", ``stepsize`` is a list. Default is 1. - gamma (float, optional): decay rate. Default is 0.1. - max_epoch (int, optional): maximum epoch (for cosine annealing). Default is 1. - - Examples:: - >>> # Decay learning rate by every 20 epochs. - >>> scheduler = torchreid.optim.build_lr_scheduler( - >>> optimizer, lr_scheduler='single_step', stepsize=20 - >>> ) - >>> # Decay learning rate at 30, 50 and 55 epochs. - >>> scheduler = torchreid.optim.build_lr_scheduler( - >>> optimizer, lr_scheduler='multi_step', stepsize=[30, 50, 55] - >>> ) - """ - if lr_scheduler not in AVAI_SCH: - raise ValueError( - 'Unsupported scheduler: {}. Must be one of {}'.format( - lr_scheduler, AVAI_SCH - ) - ) - - if lr_scheduler == 'single_step': - if isinstance(stepsize, list): - stepsize = stepsize[-1] - - if not isinstance(stepsize, int): - raise TypeError( - 'For single_step lr_scheduler, stepsize must ' - 'be an integer, but got {}'.format(type(stepsize)) - ) - - scheduler = torch.optim.lr_scheduler.StepLR( - optimizer, step_size=stepsize, gamma=gamma - ) - - elif lr_scheduler == 'multi_step': - if not isinstance(stepsize, list): - raise TypeError( - 'For multi_step lr_scheduler, stepsize must ' - 'be a list, but got {}'.format(type(stepsize)) - ) - - scheduler = torch.optim.lr_scheduler.MultiStepLR( - optimizer, milestones=stepsize, gamma=gamma - ) - - elif lr_scheduler == 'cosine': - scheduler = torch.optim.lr_scheduler.CosineAnnealingLR( - optimizer, float(max_epoch) - ) - - return scheduler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py deleted file mode 100644 index 73bb6546ea..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py +++ /dev/null @@ -1,218 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import -import warnings -import torch -import torch.nn as nn - -from .radam import RAdam - -import os - -AVAI_OPTIMS = ['adam', 'amsgrad', 'sgd', 'rmsprop', 'radam'] - - -def build_optimizer( - model, - optim='adam', - lr=0.0003, - weight_decay=5e-04, - momentum=0.9, - sgd_dampening=0, - sgd_nesterov=False, - rmsprop_alpha=0.99, - adam_beta1=0.9, - adam_beta2=0.99, - staged_lr=False, - new_layers='', - base_lr_mult=0.1 -): - """A function wrapper for building an optimizer. - - Args: - model (nn.Module): model. - optim (str, optional): optimizer. Default is "adam". - lr (float, optional): learning rate. Default is 0.0003. - weight_decay (float, optional): weight decay (L2 penalty). Default is 5e-04. - momentum (float, optional): momentum factor in sgd. Default is 0.9, - sgd_dampening (float, optional): dampening for momentum. Default is 0. - sgd_nesterov (bool, optional): enables Nesterov momentum. Default is False. - rmsprop_alpha (float, optional): smoothing constant for rmsprop. Default is 0.99. - adam_beta1 (float, optional): beta-1 value in adam. Default is 0.9. - adam_beta2 (float, optional): beta-2 value in adam. Default is 0.99, - staged_lr (bool, optional): uses different learning rates for base and new layers. Base - layers are pretrained layers while new layers are randomly initialized, e.g. the - identity classification layer. Enabling ``staged_lr`` can allow the base layers to - be trained with a smaller learning rate determined by ``base_lr_mult``, while the new - layers will take the ``lr``. Default is False. - new_layers (str or list): attribute names in ``model``. Default is empty. - base_lr_mult (float, optional): learning rate multiplier for base layers. Default is 0.1. - - Examples:: - >>> # A normal optimizer can be built by - >>> optimizer = torchreid.optim.build_optimizer(model, optim='sgd', lr=0.01) - >>> # If you want to use a smaller learning rate for pretrained layers - >>> # and the attribute name for the randomly initialized layer is 'classifier', - >>> # you can do - >>> optimizer = torchreid.optim.build_optimizer( - >>> model, optim='sgd', lr=0.01, staged_lr=True, - >>> new_layers='classifier', base_lr_mult=0.1 - >>> ) - >>> # Now the `classifier` has learning rate 0.01 but the base layers - >>> # have learning rate 0.01 * 0.1. - >>> # new_layers can also take multiple attribute names. Say the new layers - >>> # are 'fc' and 'classifier', you can do - >>> optimizer = torchreid.optim.build_optimizer( - >>> model, optim='sgd', lr=0.01, staged_lr=True, - >>> new_layers=['fc', 'classifier'], base_lr_mult=0.1 - >>> ) - """ - if optim not in AVAI_OPTIMS: - raise ValueError( - 'Unsupported optim: {}. Must be one of {}'.format( - optim, AVAI_OPTIMS - ) - ) - - if not isinstance(model, nn.Module): - raise TypeError( - 'model given to build_optimizer must be an instance of nn.Module' - ) - - if staged_lr: - if isinstance(new_layers, str): - if new_layers is None: - warnings.warn( - 'new_layers is empty, therefore, staged_lr is useless' - ) - new_layers = [new_layers] - - if isinstance(model, nn.DataParallel): - model = model.module - - base_params = [] - base_layers = [] - new_params = [] - - for name, module in model.named_children(): - if name in new_layers: - new_params += [p for p in module.parameters()] - else: - base_params += [p for p in module.parameters()] - base_layers.append(name) - - param_groups = [ - { - 'params': base_params, - 'lr': lr * base_lr_mult - }, - { - 'params': new_params - }, - ] - - else: - param_groups = model.parameters() - - if optim == 'adam': - optimizer = torch.optim.Adam( - param_groups, - lr=lr, - weight_decay=weight_decay, - betas=(adam_beta1, adam_beta2), - ) - - elif optim == 'amsgrad': - optimizer = torch.optim.Adam( - param_groups, - lr=lr, - weight_decay=weight_decay, - betas=(adam_beta1, adam_beta2), - amsgrad=True, - ) - - elif optim == 'sgd': - if os.environ['device'] == "gpu": - optimizer = torch.optim.SGD( - param_groups, - lr=lr, - momentum=momentum, - weight_decay=weight_decay, - dampening=sgd_dampening, - nesterov=sgd_nesterov, - ) - elif os.environ['device'] == "npu": - from apex.optimizers import NpuFusedSGD - optimizer = NpuFusedSGD( - param_groups, - lr=lr, - momentum=momentum, - weight_decay=weight_decay, - dampening=sgd_dampening, - nesterov=sgd_nesterov, - ) - - - elif optim == 'rmsprop': - optimizer = torch.optim.RMSprop( - param_groups, - lr=lr, - momentum=momentum, - weight_decay=weight_decay, - alpha=rmsprop_alpha, - ) - - elif optim == 'radam': - optimizer = RAdam( - param_groups, - lr=lr, - weight_decay=weight_decay, - betas=(adam_beta1, adam_beta2) - ) - - return optimizer diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py deleted file mode 100644 index 79d55b4ac8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py +++ /dev/null @@ -1,376 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. - -Imported from: https://github.com/LiyuanLucasLiu/RAdam - -Paper: https://arxiv.org/abs/1908.03265 - -@article{liu2019radam, - title={On the Variance of the Adaptive Learning Rate and Beyond}, - author={Liu, Liyuan and Jiang, Haoming and He, Pengcheng and Chen, Weizhu and Liu, Xiaodong and Gao, Jianfeng and Han, Jiawei}, - journal={arXiv preprint arXiv:1908.03265}, - year={2019} -} -""" -from __future__ import print_function, absolute_import -import math -import torch -from torch.optim.optimizer import Optimizer - - -class RAdam(Optimizer): - - def __init__( - self, - params, - lr=1e-3, - betas=(0.9, 0.999), - eps=1e-8, - weight_decay=0, - degenerated_to_sgd=True - ): - if not 0.0 <= lr: - raise ValueError("Invalid learning rate: {}".format(lr)) - if not 0.0 <= eps: - raise ValueError("Invalid epsilon value: {}".format(eps)) - if not 0.0 <= betas[0] < 1.0: - raise ValueError( - "Invalid beta parameter at index 0: {}".format(betas[0]) - ) - if not 0.0 <= betas[1] < 1.0: - raise ValueError( - "Invalid beta parameter at index 1: {}".format(betas[1]) - ) - - self.degenerated_to_sgd = degenerated_to_sgd - defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay) - self.buffer = [[None, None, None] for ind in range(10)] - super(RAdam, self).__init__(params, defaults) - - def __setstate__(self, state): - super(RAdam, self).__setstate__(state) - - def step(self, closure=None): - - loss = None - if closure is not None: - loss = closure() - - for group in self.param_groups: - - for p in group['params']: - if p.grad is None: - continue - grad = p.grad.data.float() - if grad.is_sparse: - raise RuntimeError( - 'RAdam does not support sparse gradients' - ) - - p_data_fp32 = p.data.float() - - state = self.state[p] - - if len(state) == 0: - state['step'] = 0 - state['exp_avg'] = torch.zeros_like(p_data_fp32) - state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) - else: - state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) - state['exp_avg_sq'] = state['exp_avg_sq'].type_as( - p_data_fp32 - ) - - exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] - beta1, beta2 = group['betas'] - - exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) - exp_avg.mul_(beta1).add_(1 - beta1, grad) - - state['step'] += 1 - buffered = self.buffer[int(state['step'] % 10)] - if state['step'] == buffered[0]: - N_sma, step_size = buffered[1], buffered[2] - else: - buffered[0] = state['step'] - beta2_t = beta2**state['step'] - N_sma_max = 2 / (1-beta2) - 1 - N_sma = N_sma_max - 2 * state['step' - ] * beta2_t / (1-beta2_t) - buffered[1] = N_sma - - # more conservative since it's an approximated value - if N_sma >= 5: - step_size = math.sqrt( - (1-beta2_t) * (N_sma-4) / (N_sma_max-4) * - (N_sma-2) / N_sma * N_sma_max / (N_sma_max-2) - ) / (1 - beta1**state['step']) - elif self.degenerated_to_sgd: - step_size = 1.0 / (1 - beta1**state['step']) - else: - step_size = -1 - buffered[2] = step_size - - # more conservative since it's an approximated value - if N_sma >= 5: - if group['weight_decay'] != 0: - p_data_fp32.add_( - -group['weight_decay'] * group['lr'], p_data_fp32 - ) - denom = exp_avg_sq.sqrt().add_(group['eps']) - p_data_fp32.addcdiv_( - -step_size * group['lr'], exp_avg, denom - ) - p.data.copy_(p_data_fp32) - elif step_size > 0: - if group['weight_decay'] != 0: - p_data_fp32.add_( - -group['weight_decay'] * group['lr'], p_data_fp32 - ) - p_data_fp32.add_(-step_size * group['lr'], exp_avg) - p.data.copy_(p_data_fp32) - - return loss - - -class PlainRAdam(Optimizer): - - def __init__( - self, - params, - lr=1e-3, - betas=(0.9, 0.999), - eps=1e-8, - weight_decay=0, - degenerated_to_sgd=True - ): - if not 0.0 <= lr: - raise ValueError("Invalid learning rate: {}".format(lr)) - if not 0.0 <= eps: - raise ValueError("Invalid epsilon value: {}".format(eps)) - if not 0.0 <= betas[0] < 1.0: - raise ValueError( - "Invalid beta parameter at index 0: {}".format(betas[0]) - ) - if not 0.0 <= betas[1] < 1.0: - raise ValueError( - "Invalid beta parameter at index 1: {}".format(betas[1]) - ) - - self.degenerated_to_sgd = degenerated_to_sgd - defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay) - - super(PlainRAdam, self).__init__(params, defaults) - - def __setstate__(self, state): - super(PlainRAdam, self).__setstate__(state) - - def step(self, closure=None): - - loss = None - if closure is not None: - loss = closure() - - for group in self.param_groups: - - for p in group['params']: - if p.grad is None: - continue - grad = p.grad.data.float() - if grad.is_sparse: - raise RuntimeError( - 'RAdam does not support sparse gradients' - ) - - p_data_fp32 = p.data.float() - - state = self.state[p] - - if len(state) == 0: - state['step'] = 0 - state['exp_avg'] = torch.zeros_like(p_data_fp32) - state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) - else: - state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) - state['exp_avg_sq'] = state['exp_avg_sq'].type_as( - p_data_fp32 - ) - - exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] - beta1, beta2 = group['betas'] - - exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) - exp_avg.mul_(beta1).add_(1 - beta1, grad) - - state['step'] += 1 - beta2_t = beta2**state['step'] - N_sma_max = 2 / (1-beta2) - 1 - N_sma = N_sma_max - 2 * state['step'] * beta2_t / (1-beta2_t) - - # more conservative since it's an approximated value - if N_sma >= 5: - if group['weight_decay'] != 0: - p_data_fp32.add_( - -group['weight_decay'] * group['lr'], p_data_fp32 - ) - step_size = group['lr'] * math.sqrt( - (1-beta2_t) * (N_sma-4) / (N_sma_max-4) * - (N_sma-2) / N_sma * N_sma_max / (N_sma_max-2) - ) / (1 - beta1**state['step']) - denom = exp_avg_sq.sqrt().add_(group['eps']) - p_data_fp32.addcdiv_(-step_size, exp_avg, denom) - p.data.copy_(p_data_fp32) - elif self.degenerated_to_sgd: - if group['weight_decay'] != 0: - p_data_fp32.add_( - -group['weight_decay'] * group['lr'], p_data_fp32 - ) - step_size = group['lr'] / (1 - beta1**state['step']) - p_data_fp32.add_(-step_size, exp_avg) - p.data.copy_(p_data_fp32) - - return loss - - -class AdamW(Optimizer): - - def __init__( - self, - params, - lr=1e-3, - betas=(0.9, 0.999), - eps=1e-8, - weight_decay=0, - warmup=0 - ): - if not 0.0 <= lr: - raise ValueError("Invalid learning rate: {}".format(lr)) - if not 0.0 <= eps: - raise ValueError("Invalid epsilon value: {}".format(eps)) - if not 0.0 <= betas[0] < 1.0: - raise ValueError( - "Invalid beta parameter at index 0: {}".format(betas[0]) - ) - if not 0.0 <= betas[1] < 1.0: - raise ValueError( - "Invalid beta parameter at index 1: {}".format(betas[1]) - ) - - defaults = dict( - lr=lr, - betas=betas, - eps=eps, - weight_decay=weight_decay, - warmup=warmup - ) - super(AdamW, self).__init__(params, defaults) - - def __setstate__(self, state): - super(AdamW, self).__setstate__(state) - - def step(self, closure=None): - loss = None - if closure is not None: - loss = closure() - - for group in self.param_groups: - - for p in group['params']: - if p.grad is None: - continue - grad = p.grad.data.float() - if grad.is_sparse: - raise RuntimeError( - 'Adam does not support sparse gradients, please consider SparseAdam instead' - ) - - p_data_fp32 = p.data.float() - - state = self.state[p] - - if len(state) == 0: - state['step'] = 0 - state['exp_avg'] = torch.zeros_like(p_data_fp32) - state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) - else: - state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) - state['exp_avg_sq'] = state['exp_avg_sq'].type_as( - p_data_fp32 - ) - - exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] - beta1, beta2 = group['betas'] - - state['step'] += 1 - - exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) - exp_avg.mul_(beta1).add_(1 - beta1, grad) - - denom = exp_avg_sq.sqrt().add_(group['eps']) - bias_correction1 = 1 - beta1**state['step'] - bias_correction2 = 1 - beta2**state['step'] - - if group['warmup'] > state['step']: - scheduled_lr = 1e-8 + state['step'] * group['lr'] / group[ - 'warmup'] - else: - scheduled_lr = group['lr'] - - step_size = scheduled_lr * math.sqrt( - bias_correction2 - ) / bias_correction1 - - if group['weight_decay'] != 0: - p_data_fp32.add_( - -group['weight_decay'] * scheduled_lr, p_data_fp32 - ) - - p_data_fp32.addcdiv_(-step_size, exp_avg, denom) - - p.data.copy_(p_data_fp32) - - return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py deleted file mode 100644 index 8f39046e67..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py +++ /dev/null @@ -1,57 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .tools import * -from .rerank import re_ranking -from .loggers import * -from .avgmeter import * -from .reidtools import * -from .torchtools import * -from .model_complexity import compute_model_complexity -from .feature_extractor import FeatureExtractor diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py deleted file mode 100644 index 82ef00ed66..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py +++ /dev/null @@ -1,120 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -from collections import defaultdict -import torch - -__all__ = ['AverageMeter', 'MetricMeter'] - - -class AverageMeter(object): - """Computes and stores the average and current value. - - Examples:: - >>> # Initialize a meter to record loss - >>> losses = AverageMeter() - >>> # Update meter after every minibatch update - >>> losses.update(loss_value, batch_size) - """ - - def __init__(self): - self.reset() - - def reset(self): - self.val = 0 - self.avg = 0 - self.sum = 0 - self.count = 0 - - def update(self, val, n=1): - self.val = val - self.sum += val * n - self.count += n - self.avg = self.sum / self.count - - -class MetricMeter(object): - """A collection of metrics. - - Source: https://github.com/KaiyangZhou/Dassl.pytorch - - Examples:: - >>> # 1. Create an instance of MetricMeter - >>> metric = MetricMeter() - >>> # 2. Update using a dictionary as input - >>> input_dict = {'loss_1': value_1, 'loss_2': value_2} - >>> metric.update(input_dict) - >>> # 3. Convert to string and print - >>> print(str(metric)) - """ - - def __init__(self, delimiter='\t'): - self.meters = defaultdict(AverageMeter) - self.delimiter = delimiter - - def update(self, input_dict): - if input_dict is None: - return - - if not isinstance(input_dict, dict): - raise TypeError( - 'Input to MetricMeter.update() must be a dictionary' - ) - - for k, v in input_dict.items(): - if isinstance(v, torch.Tensor): - v = v.item() - self.meters[k].update(v) - - def __str__(self): - output_str = [] - for name, meter in self.meters.items(): - output_str.append( - '{} {:.4f} ({:.4f})'.format(name, meter.val, meter.avg) - ) - return self.delimiter.join(output_str) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py deleted file mode 100644 index b25cc9959d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py +++ /dev/null @@ -1,199 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import -import numpy as np -import torch -import torchvision.transforms as T -from PIL import Image - -from torchreid.utils import ( - check_isfile, load_pretrained_weights, compute_model_complexity -) -from torchreid.models import build_model - - -class FeatureExtractor(object): - """A simple API for feature extraction. - - FeatureExtractor can be used like a python function, which - accepts input of the following types: - - a list of strings (image paths) - - a list of numpy.ndarray each with shape (H, W, C) - - a single string (image path) - - a single numpy.ndarray with shape (H, W, C) - - a torch.Tensor with shape (B, C, H, W) or (C, H, W) - - Returned is a torch tensor with shape (B, D) where D is the - feature dimension. - - Args: - model_name (str): model name. - model_path (str): path to model weights. - image_size (sequence or int): image height and width. - pixel_mean (list): pixel mean for normalization. - pixel_std (list): pixel std for normalization. - pixel_norm (bool): whether to normalize pixels. - device (str): 'cpu' or 'cuda' (could be specific gpu devices). - verbose (bool): show model details. - - Examples:: - - from torchreid.utils import FeatureExtractor - - extractor = FeatureExtractor( - model_name='osnet_x1_0', - model_path='a/b/c/model.pth.tar', - device='cuda' - ) - - image_list = [ - 'a/b/c/image001.jpg', - 'a/b/c/image002.jpg', - 'a/b/c/image003.jpg', - 'a/b/c/image004.jpg', - 'a/b/c/image005.jpg' - ] - - features = extractor(image_list) - print(features.shape) # output (5, 512) - """ - - def __init__( - self, - model_name='', - model_path='', - image_size=(256, 128), - pixel_mean=[0.485, 0.456, 0.406], - pixel_std=[0.229, 0.224, 0.225], - pixel_norm=True, - device='cuda', - verbose=True - ): - # Build model - model = build_model( - model_name, - num_classes=1, - pretrained=True, - use_gpu=device.startswith('cuda') - ) - model.eval() - - if verbose: - num_params, flops = compute_model_complexity( - model, (1, 3, image_size[0], image_size[1]) - ) - print('Model: {}'.format(model_name)) - print('- params: {:,}'.format(num_params)) - print('- flops: {:,}'.format(flops)) - - if model_path and check_isfile(model_path): - load_pretrained_weights(model, model_path) - - # Build transform functions - transforms = [] - transforms += [T.Resize(image_size)] - transforms += [T.ToTensor()] - if pixel_norm: - transforms += [T.Normalize(mean=pixel_mean, std=pixel_std)] - preprocess = T.Compose(transforms) - - to_pil = T.ToPILImage() - - device = torch.device(device) - model.to(device) - - # Class attributes - self.model = model - self.preprocess = preprocess - self.to_pil = to_pil - self.device = device - - def __call__(self, input): - if isinstance(input, list): - images = [] - - for element in input: - if isinstance(element, str): - image = Image.open(element).convert('RGB') - - elif isinstance(element, np.ndarray): - image = self.to_pil(element) - - else: - raise TypeError( - 'Type of each element must belong to [str | numpy.ndarray]' - ) - - image = self.preprocess(image) - images.append(image) - - images = torch.stack(images, dim=0) - images = images.to(self.device) - - elif isinstance(input, str): - image = Image.open(input).convert('RGB') - image = self.preprocess(image) - images = image.unsqueeze(0).to(self.device) - - elif isinstance(input, np.ndarray): - image = self.to_pil(input) - image = self.preprocess(image) - images = image.unsqueeze(0).to(self.device) - - elif isinstance(input, torch.Tensor): - if input.dim() == 3: - input = input.unsqueeze(0) - images = input.to(self.device) - - else: - raise NotImplementedError - - with torch.no_grad(): - features = self.model(images) - - return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py deleted file mode 100644 index f61ad5cd91..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py +++ /dev/null @@ -1,193 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import -import os -import sys -import os.path as osp - -from .tools import mkdir_if_missing - -__all__ = ['Logger', 'RankLogger'] - - -class Logger(object): - """Writes console output to external text file. - - Imported from ``_ - - Args: - fpath (str): directory to save logging file. - - Examples:: - >>> import sys - >>> import os - >>> import os.path as osp - >>> from torchreid.utils import Logger - >>> save_dir = 'log/resnet50-softmax-market1501' - >>> log_name = 'train.log' - >>> sys.stdout = Logger(osp.join(args.save_dir, log_name)) - """ - - def __init__(self, fpath=None): - self.console = sys.stdout - self.file = None - if fpath is not None: - mkdir_if_missing(osp.dirname(fpath)) - self.file = open(fpath, 'w') - - def __del__(self): - self.close() - - def __enter__(self): - pass - - def __exit__(self, *args): - self.close() - - def write(self, msg): - self.console.write(msg) - if self.file is not None: - self.file.write(msg) - - def flush(self): - self.console.flush() - if self.file is not None: - self.file.flush() - os.fsync(self.file.fileno()) - - def close(self): - self.console.close() - if self.file is not None: - self.file.close() - - -class RankLogger(object): - """Records the rank1 matching accuracy obtained for each - test dataset at specified evaluation steps and provides a function - to show the summarized results, which are convenient for analysis. - - Args: - sources (str or list): source dataset name(s). - targets (str or list): target dataset name(s). - - Examples:: - >>> from torchreid.utils import RankLogger - >>> s = 'market1501' - >>> t = 'market1501' - >>> ranklogger = RankLogger(s, t) - >>> ranklogger.write(t, 10, 0.5) - >>> ranklogger.write(t, 20, 0.7) - >>> ranklogger.write(t, 30, 0.9) - >>> ranklogger.show_summary() - >>> # You will see: - >>> # => Show performance summary - >>> # market1501 (source) - >>> # - epoch 10 rank1 50.0% - >>> # - epoch 20 rank1 70.0% - >>> # - epoch 30 rank1 90.0% - >>> # If there are multiple test datasets - >>> t = ['market1501', 'dukemtmcreid'] - >>> ranklogger = RankLogger(s, t) - >>> ranklogger.write(t[0], 10, 0.5) - >>> ranklogger.write(t[0], 20, 0.7) - >>> ranklogger.write(t[0], 30, 0.9) - >>> ranklogger.write(t[1], 10, 0.1) - >>> ranklogger.write(t[1], 20, 0.2) - >>> ranklogger.write(t[1], 30, 0.3) - >>> ranklogger.show_summary() - >>> # You can see: - >>> # => Show performance summary - >>> # market1501 (source) - >>> # - epoch 10 rank1 50.0% - >>> # - epoch 20 rank1 70.0% - >>> # - epoch 30 rank1 90.0% - >>> # dukemtmcreid (target) - >>> # - epoch 10 rank1 10.0% - >>> # - epoch 20 rank1 20.0% - >>> # - epoch 30 rank1 30.0% - """ - - def __init__(self, sources, targets): - self.sources = sources - self.targets = targets - - if isinstance(self.sources, str): - self.sources = [self.sources] - - if isinstance(self.targets, str): - self.targets = [self.targets] - - self.logger = { - name: { - 'epoch': [], - 'rank1': [] - } - for name in self.targets - } - - def write(self, name, epoch, rank1): - """Writes result. - - Args: - name (str): dataset name. - epoch (int): current epoch. - rank1 (float): rank1 result. - """ - self.logger[name]['epoch'].append(epoch) - self.logger[name]['rank1'].append(rank1) - - def show_summary(self): - """Shows saved results.""" - print('=> Show performance summary') - for name in self.targets: - from_where = 'source' if name in self.sources else 'target' - print('{} ({})'.format(name, from_where)) - for epoch, rank1 in zip( - self.logger[name]['epoch'], self.logger[name]['rank1'] - ): - print('- epoch {}\t rank1 {:.1%}'.format(epoch, rank1)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py deleted file mode 100644 index af673038e8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py +++ /dev/null @@ -1,410 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import math -import numpy as np -from itertools import repeat -from collections import namedtuple, defaultdict -import torch - -__all__ = ['compute_model_complexity'] -""" -Utility -""" - - -def _ntuple(n): - - def parse(x): - if isinstance(x, int): - return tuple(repeat(x, n)) - return x - - return parse - - -_single = _ntuple(1) -_pair = _ntuple(2) -_triple = _ntuple(3) -""" -Convolution -""" - - -def hook_convNd(m, x, y): - k = torch.prod(torch.Tensor(m.kernel_size)).item() - cin = m.in_channels - flops_per_ele = k * cin # + (k*cin-1) - if m.bias is not None: - flops_per_ele += 1 - flops = flops_per_ele * y.numel() / m.groups - return int(flops) - - -""" -Pooling -""" - - -def hook_maxpool1d(m, x, y): - flops_per_ele = m.kernel_size - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_maxpool2d(m, x, y): - k = _pair(m.kernel_size) - k = torch.prod(torch.Tensor(k)).item() - # ops: compare - flops_per_ele = k - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_maxpool3d(m, x, y): - k = _triple(m.kernel_size) - k = torch.prod(torch.Tensor(k)).item() - flops_per_ele = k - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_avgpool1d(m, x, y): - flops_per_ele = m.kernel_size - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_avgpool2d(m, x, y): - k = _pair(m.kernel_size) - k = torch.prod(torch.Tensor(k)).item() - flops_per_ele = k - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_avgpool3d(m, x, y): - k = _triple(m.kernel_size) - k = torch.prod(torch.Tensor(k)).item() - flops_per_ele = k - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapmaxpool1d(m, x, y): - x = x[0] - out_size = m.output_size - k = math.ceil(x.size(2) / out_size) - flops_per_ele = k - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapmaxpool2d(m, x, y): - x = x[0] - out_size = _pair(m.output_size) - k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) - k = torch.prod(torch.ceil(k)).item() - flops_per_ele = k - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapmaxpool3d(m, x, y): - x = x[0] - out_size = _triple(m.output_size) - k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) - k = torch.prod(torch.ceil(k)).item() - flops_per_ele = k - 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapavgpool1d(m, x, y): - x = x[0] - out_size = m.output_size - k = math.ceil(x.size(2) / out_size) - flops_per_ele = k - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapavgpool2d(m, x, y): - x = x[0] - out_size = _pair(m.output_size) - k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) - k = torch.prod(torch.ceil(k)).item() - flops_per_ele = k - flops = flops_per_ele * y.numel() - return int(flops) - - -def hook_adapavgpool3d(m, x, y): - x = x[0] - out_size = _triple(m.output_size) - k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) - k = torch.prod(torch.ceil(k)).item() - flops_per_ele = k - flops = flops_per_ele * y.numel() - return int(flops) - - -""" -Non-linear activations -""" - - -def hook_relu(m, x, y): - # eq: max(0, x) - num_ele = y.numel() - return int(num_ele) - - -def hook_leakyrelu(m, x, y): - # eq: max(0, x) + negative_slope*min(0, x) - num_ele = y.numel() - flops = 3 * num_ele - return int(flops) - - -""" -Normalization -""" - - -def hook_batchnormNd(m, x, y): - num_ele = y.numel() - flops = 2 * num_ele # mean and std - if m.affine: - flops += 2 * num_ele # gamma and beta - return int(flops) - - -def hook_instancenormNd(m, x, y): - return hook_batchnormNd(m, x, y) - - -def hook_groupnorm(m, x, y): - return hook_batchnormNd(m, x, y) - - -def hook_layernorm(m, x, y): - num_ele = y.numel() - flops = 2 * num_ele # mean and std - if m.elementwise_affine: - flops += 2 * num_ele # gamma and beta - return int(flops) - - -""" -Linear -""" - - -def hook_linear(m, x, y): - flops_per_ele = m.in_features # + (m.in_features-1) - if m.bias is not None: - flops_per_ele += 1 - flops = flops_per_ele * y.numel() - return int(flops) - - -__generic_flops_counter = { - # Convolution - 'Conv1d': hook_convNd, - 'Conv2d': hook_convNd, - 'Conv3d': hook_convNd, - # Pooling - 'MaxPool1d': hook_maxpool1d, - 'MaxPool2d': hook_maxpool2d, - 'MaxPool3d': hook_maxpool3d, - 'AvgPool1d': hook_avgpool1d, - 'AvgPool2d': hook_avgpool2d, - 'AvgPool3d': hook_avgpool3d, - 'AdaptiveMaxPool1d': hook_adapmaxpool1d, - 'AdaptiveMaxPool2d': hook_adapmaxpool2d, - 'AdaptiveMaxPool3d': hook_adapmaxpool3d, - 'AdaptiveAvgPool1d': hook_adapavgpool1d, - 'AdaptiveAvgPool2d': hook_adapavgpool2d, - 'AdaptiveAvgPool3d': hook_adapavgpool3d, - # Non-linear activations - 'ReLU': hook_relu, - 'ReLU6': hook_relu, - 'LeakyReLU': hook_leakyrelu, - # Normalization - 'BatchNorm1d': hook_batchnormNd, - 'BatchNorm2d': hook_batchnormNd, - 'BatchNorm3d': hook_batchnormNd, - 'InstanceNorm1d': hook_instancenormNd, - 'InstanceNorm2d': hook_instancenormNd, - 'InstanceNorm3d': hook_instancenormNd, - 'GroupNorm': hook_groupnorm, - 'LayerNorm': hook_layernorm, - # Linear - 'Linear': hook_linear, -} - -__conv_linear_flops_counter = { - # Convolution - 'Conv1d': hook_convNd, - 'Conv2d': hook_convNd, - 'Conv3d': hook_convNd, - # Linear - 'Linear': hook_linear, -} - - -def _get_flops_counter(only_conv_linear): - if only_conv_linear: - return __conv_linear_flops_counter - return __generic_flops_counter - - -def compute_model_complexity( - model, input_size, verbose=False, only_conv_linear=True -): - """Returns number of parameters and FLOPs. - - .. note:: - (1) this function only provides an estimate of the theoretical time complexity - rather than the actual running time which depends on implementations and hardware, - and (2) the FLOPs is only counted for layers that are used at test time. This means - that redundant layers such as person ID classification layer will be ignored as it - is discarded when doing feature extraction. Note that the inference graph depends on - how you construct the computations in ``forward()``. - - Args: - model (nn.Module): network model. - input_size (tuple): input size, e.g. (1, 3, 256, 128). - verbose (bool, optional): shows detailed complexity of - each module. Default is False. - only_conv_linear (bool, optional): only considers convolution - and linear layers when counting flops. Default is True. - If set to False, flops of all layers will be counted. - - Examples:: - >>> from torchreid import models, utils - >>> model = models.build_model(name='resnet50', num_classes=1000) - >>> num_params, flops = utils.compute_model_complexity(model, (1, 3, 256, 128), verbose=True) - """ - registered_handles = [] - layer_list = [] - layer = namedtuple('layer', ['class_name', 'params', 'flops']) - - def _add_hooks(m): - - def _has_submodule(m): - return len(list(m.children())) > 0 - - def _hook(m, x, y): - params = sum(p.numel() for p in m.parameters()) - class_name = str(m.__class__.__name__) - flops_counter = _get_flops_counter(only_conv_linear) - if class_name in flops_counter: - flops = flops_counter[class_name](m, x, y) - else: - flops = 0 - layer_list.append( - layer(class_name=class_name, params=params, flops=flops) - ) - - # only consider the very basic nn layer - if _has_submodule(m): - return - - handle = m.register_forward_hook(_hook) - registered_handles.append(handle) - - default_train_mode = model.training - - model.eval().apply(_add_hooks) - input = torch.rand(input_size) - if next(model.parameters()).is_cuda: - input = input.cuda() - model(input) # forward - - for handle in registered_handles: - handle.remove() - - model.train(default_train_mode) - - if verbose: - per_module_params = defaultdict(list) - per_module_flops = defaultdict(list) - - total_params, total_flops = 0, 0 - - for layer in layer_list: - total_params += layer.params - total_flops += layer.flops - if verbose: - per_module_params[layer.class_name].append(layer.params) - per_module_flops[layer.class_name].append(layer.flops) - - if verbose: - num_udscore = 55 - print(' {}'.format('-' * num_udscore)) - print(' Model complexity with input size {}'.format(input_size)) - print(' {}'.format('-' * num_udscore)) - for class_name in per_module_params: - params = int(np.sum(per_module_params[class_name])) - flops = int(np.sum(per_module_flops[class_name])) - print( - ' {} (params={:,}, flops={:,})'.format( - class_name, params, flops - ) - ) - print(' {}'.format('-' * num_udscore)) - print( - ' Total (params={:,}, flops={:,})'.format( - total_params, total_flops - ) - ) - print(' {}'.format('-' * num_udscore)) - - return total_params, total_flops diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py deleted file mode 100644 index 4cf1a7ae74..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py +++ /dev/null @@ -1,201 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import -import numpy as np -import shutil -import os.path as osp -import cv2 - -from .tools import mkdir_if_missing - -__all__ = ['visualize_ranked_results'] - -GRID_SPACING = 10 -QUERY_EXTRA_SPACING = 90 -BW = 5 # border width -GREEN = (0, 255, 0) -RED = (0, 0, 255) - - -def visualize_ranked_results( - distmat, dataset, data_type, width=128, height=256, save_dir='', topk=10 -): - """Visualizes ranked results. - - Supports both image-reid and video-reid. - - For image-reid, ranks will be plotted in a single figure. For video-reid, ranks will be - saved in folders each containing a tracklet. - - Args: - distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). - dataset (tuple): a 2-tuple containing (query, gallery), each of which contains - tuples of (img_path(s), pid, camid, dsetid). - data_type (str): "image" or "video". - width (int, optional): resized image width. Default is 128. - height (int, optional): resized image height. Default is 256. - save_dir (str): directory to save output images. - topk (int, optional): denoting top-k images in the rank list to be visualized. - Default is 10. - """ - num_q, num_g = distmat.shape - mkdir_if_missing(save_dir) - - print('# query: {}\n# gallery {}'.format(num_q, num_g)) - print('Visualizing top-{} ranks ...'.format(topk)) - - query, gallery = dataset - assert num_q == len(query) - assert num_g == len(gallery) - - indices = np.argsort(distmat, axis=1) - - def _cp_img_to(src, dst, rank, prefix, matched=False): - """ - Args: - src: image path or tuple (for vidreid) - dst: target directory - rank: int, denoting ranked position, starting from 1 - prefix: string - matched: bool - """ - if isinstance(src, (tuple, list)): - if prefix == 'gallery': - suffix = 'TRUE' if matched else 'FALSE' - dst = osp.join( - dst, prefix + '_top' + str(rank).zfill(3) - ) + '_' + suffix - else: - dst = osp.join(dst, prefix + '_top' + str(rank).zfill(3)) - mkdir_if_missing(dst) - for img_path in src: - shutil.copy(img_path, dst) - else: - dst = osp.join( - dst, prefix + '_top' + str(rank).zfill(3) + '_name_' + - osp.basename(src) - ) - shutil.copy(src, dst) - - for q_idx in range(num_q): - qimg_path, qpid, qcamid = query[q_idx][:3] - qimg_path_name = qimg_path[0] if isinstance( - qimg_path, (tuple, list) - ) else qimg_path - - if data_type == 'image': - qimg = cv2.imread(qimg_path) - qimg = cv2.resize(qimg, (width, height)) - qimg = cv2.copyMakeBorder( - qimg, BW, BW, BW, BW, cv2.BORDER_CONSTANT, value=(0, 0, 0) - ) - # resize twice to ensure that the border width is consistent across images - qimg = cv2.resize(qimg, (width, height)) - num_cols = topk + 1 - grid_img = 255 * np.ones( - ( - height, - num_cols*width + topk*GRID_SPACING + QUERY_EXTRA_SPACING, 3 - ), - dtype=np.uint8 - ) - grid_img[:, :width, :] = qimg - else: - qdir = osp.join( - save_dir, osp.basename(osp.splitext(qimg_path_name)[0]) - ) - mkdir_if_missing(qdir) - _cp_img_to(qimg_path, qdir, rank=0, prefix='query') - - rank_idx = 1 - for g_idx in indices[q_idx, :]: - gimg_path, gpid, gcamid = gallery[g_idx][:3] - invalid = (qpid == gpid) & (qcamid == gcamid) - - if not invalid: - matched = gpid == qpid - if data_type == 'image': - border_color = GREEN if matched else RED - gimg = cv2.imread(gimg_path) - gimg = cv2.resize(gimg, (width, height)) - gimg = cv2.copyMakeBorder( - gimg, - BW, - BW, - BW, - BW, - cv2.BORDER_CONSTANT, - value=border_color - ) - gimg = cv2.resize(gimg, (width, height)) - start = rank_idx*width + rank_idx*GRID_SPACING + QUERY_EXTRA_SPACING - end = ( - rank_idx+1 - ) * width + rank_idx*GRID_SPACING + QUERY_EXTRA_SPACING - grid_img[:, start:end, :] = gimg - else: - _cp_img_to( - gimg_path, - qdir, - rank=rank_idx, - prefix='gallery', - matched=matched - ) - - rank_idx += 1 - if rank_idx > topk: - break - - if data_type == 'image': - imname = osp.basename(osp.splitext(qimg_path_name)[0]) - cv2.imwrite(osp.join(save_dir, imname + '.jpg'), grid_img) - - if (q_idx+1) % 100 == 0: - print('- done {}/{}'.format(q_idx + 1, num_q)) - - print('Done. Images have been saved to "{}" ...'.format(save_dir)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py deleted file mode 100644 index f3e0ab8faf..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py +++ /dev/null @@ -1,157 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. - -Source: https://github.com/zhunzhong07/person-re-ranking - -Created on Mon Jun 26 14:46:56 2017 -@author: luohao -Modified by Houjing Huang, 2017-12-22. -- This version accepts distance matrix instead of raw features. -- The difference of `/` division between python 2 and 3 is handled. -- numpy.float16 is replaced by numpy.float32 for numerical precision. - -CVPR2017 paper:Zhong Z, Zheng L, Cao D, et al. Re-ranking Person Re-identification with k-reciprocal Encoding[J]. 2017. -url:http://openaccess.thecvf.com/content_cvpr_2017/papers/Zhong_Re-Ranking_Person_Re-Identification_CVPR_2017_paper.pdf -Matlab version: https://github.com/zhunzhong07/person-re-ranking - -API -q_g_dist: query-gallery distance matrix, numpy array, shape [num_query, num_gallery] -q_q_dist: query-query distance matrix, numpy array, shape [num_query, num_query] -g_g_dist: gallery-gallery distance matrix, numpy array, shape [num_gallery, num_gallery] -k1, k2, lambda_value: parameters, the original paper is (k1=20, k2=6, lambda_value=0.3) -Returns: - final_dist: re-ranked distance, numpy array, shape [num_query, num_gallery] -""" -from __future__ import division, print_function, absolute_import -import numpy as np - -__all__ = ['re_ranking'] - - -def re_ranking(q_g_dist, q_q_dist, g_g_dist, k1=20, k2=6, lambda_value=0.3): - - # The following naming, e.g. gallery_num, is different from outer scope. - # Don't care about it. - - original_dist = np.concatenate( - [ - np.concatenate([q_q_dist, q_g_dist], axis=1), - np.concatenate([q_g_dist.T, g_g_dist], axis=1) - ], - axis=0 - ) - original_dist = np.power(original_dist, 2).astype(np.float32) - original_dist = np.transpose( - 1. * original_dist / np.max(original_dist, axis=0) - ) - V = np.zeros_like(original_dist).astype(np.float32) - initial_rank = np.argsort(original_dist).astype(np.int32) - - query_num = q_g_dist.shape[0] - gallery_num = q_g_dist.shape[0] + q_g_dist.shape[1] - all_num = gallery_num - - for i in range(all_num): - # k-reciprocal neighbors - forward_k_neigh_index = initial_rank[i, :k1 + 1] - backward_k_neigh_index = initial_rank[forward_k_neigh_index, :k1 + 1] - fi = np.where(backward_k_neigh_index == i)[0] - k_reciprocal_index = forward_k_neigh_index[fi] - k_reciprocal_expansion_index = k_reciprocal_index - for j in range(len(k_reciprocal_index)): - candidate = k_reciprocal_index[j] - candidate_forward_k_neigh_index = initial_rank[ - candidate, :int(np.around(k1 / 2.)) + 1] - candidate_backward_k_neigh_index = initial_rank[ - candidate_forward_k_neigh_index, :int(np.around(k1 / 2.)) + 1] - fi_candidate = np.where( - candidate_backward_k_neigh_index == candidate - )[0] - candidate_k_reciprocal_index = candidate_forward_k_neigh_index[ - fi_candidate] - if len( - np. - intersect1d(candidate_k_reciprocal_index, k_reciprocal_index) - ) > 2. / 3 * len(candidate_k_reciprocal_index): - k_reciprocal_expansion_index = np.append( - k_reciprocal_expansion_index, candidate_k_reciprocal_index - ) - - k_reciprocal_expansion_index = np.unique(k_reciprocal_expansion_index) - weight = np.exp(-original_dist[i, k_reciprocal_expansion_index]) - V[i, k_reciprocal_expansion_index] = 1. * weight / np.sum(weight) - original_dist = original_dist[:query_num, ] - if k2 != 1: - V_qe = np.zeros_like(V, dtype=np.float32) - for i in range(all_num): - V_qe[i, :] = np.mean(V[initial_rank[i, :k2], :], axis=0) - V = V_qe - del V_qe - del initial_rank - invIndex = [] - for i in range(gallery_num): - invIndex.append(np.where(V[:, i] != 0)[0]) - - jaccard_dist = np.zeros_like(original_dist, dtype=np.float32) - - for i in range(query_num): - temp_min = np.zeros(shape=[1, gallery_num], dtype=np.float32) - indNonZero = np.where(V[i, :] != 0)[0] - indImages = [] - indImages = [invIndex[ind] for ind in indNonZero] - for j in range(len(indNonZero)): - temp_min[0, indImages[j]] = temp_min[0, indImages[j]] + np.minimum( - V[i, indNonZero[j]], V[indImages[j], indNonZero[j]] - ) - jaccard_dist[i] = 1 - temp_min / (2.-temp_min) - - final_dist = jaccard_dist * (1-lambda_value) + original_dist*lambda_value - del original_dist - del V - del jaccard_dist - final_dist = final_dist[:query_num, query_num:] - return final_dist diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py deleted file mode 100644 index d30752e112..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py +++ /dev/null @@ -1,189 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import os -import sys -import json -import time -import errno -import numpy as np -import random -import os.path as osp -import warnings -import PIL -import torch -from PIL import Image - -__all__ = [ - 'mkdir_if_missing', 'check_isfile', 'read_json', 'write_json', - 'set_random_seed', 'download_url', 'read_image', 'collect_env_info', - 'listdir_nohidden' -] - - -def mkdir_if_missing(dirname): - """Creates dirname if it is missing.""" - if not osp.exists(dirname): - try: - os.makedirs(dirname) - except OSError as e: - if e.errno != errno.EEXIST: - raise - - -def check_isfile(fpath): - """Checks if the given path is a file. - - Args: - fpath (str): file path. - - Returns: - bool - """ - isfile = osp.isfile(fpath) - if not isfile: - warnings.warn('No file found at "{}"'.format(fpath)) - return isfile - - -def read_json(fpath): - """Reads json file from a path.""" - with open(fpath, 'r') as f: - obj = json.load(f) - return obj - - -def write_json(obj, fpath): - """Writes to a json file.""" - mkdir_if_missing(osp.dirname(fpath)) - with open(fpath, 'w') as f: - json.dump(obj, f, indent=4, separators=(',', ': ')) - - -def set_random_seed(seed): - random.seed(seed) - np.random.seed(seed) - torch.manual_seed(seed) - - -def download_url(url, dst): - """Downloads file from a url to a destination. - - Args: - url (str): url to download file. - dst (str): destination path. - """ - from six.moves import urllib - print('* url="{}"'.format(url)) - print('* destination="{}"'.format(dst)) - - def _reporthook(count, block_size, total_size): - global start_time - if count == 0: - start_time = time.time() - return - duration = time.time() - start_time - progress_size = int(count * block_size) - speed = int(progress_size / (1024*duration)) - percent = int(count * block_size * 100 / total_size) - sys.stdout.write( - '\r...%d%%, %d MB, %d KB/s, %d seconds passed' % - (percent, progress_size / (1024*1024), speed, duration) - ) - sys.stdout.flush() - - urllib.request.urlretrieve(url, dst, _reporthook) - sys.stdout.write('\n') - - -def read_image(path): - """Reads image from path using ``PIL.Image``. - - Args: - path (str): path to an image. - - Returns: - PIL image - """ - got_img = False - if not osp.exists(path): - raise IOError('"{}" does not exist'.format(path)) - while not got_img: - try: - img = Image.open(path).convert('RGB') - got_img = True - except IOError: - print( - 'IOError incurred when reading "{}". Will redo. Don\'t worry. Just chill.' - .format(path) - ) - return img - - -def collect_env_info(): - """Returns env info as a string. - - Code source: github.com/facebookresearch/maskrcnn-benchmark - """ - from torch.utils.collect_env import get_pretty_env_info - env_str = get_pretty_env_info() - env_str += '\n Pillow ({})'.format(PIL.__version__) - return env_str - - -def listdir_nohidden(path, sort=False): - """List non-hidden items in a directory. - - Args: - path (str): directory path. - sort (bool): sort the items. - """ - items = [f for f in os.listdir(path) if not f.startswith('.')] - if sort: - items.sort() - return items diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py deleted file mode 100644 index f9d5d6f12d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py +++ /dev/null @@ -1,363 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import pickle -import shutil -import os.path as osp -import warnings -from functools import partial -from collections import OrderedDict -import torch -import torch.nn as nn - -from .tools import mkdir_if_missing - -__all__ = [ - 'save_checkpoint', 'load_checkpoint', 'resume_from_checkpoint', - 'open_all_layers', 'open_specified_layers', 'count_num_param', - 'load_pretrained_weights' -] - - -def save_checkpoint( - state, save_dir, is_best=False, remove_module_from_keys=False -): - r"""Saves checkpoint. - - Args: - state (dict): dictionary. - save_dir (str): directory to save checkpoint. - is_best (bool, optional): if True, this checkpoint will be copied and named - ``model-best.pth.tar``. Default is False. - remove_module_from_keys (bool, optional): whether to remove "module." - from layer names. Default is False. - - Examples:: - >>> state = { - >>> 'state_dict': model.state_dict(), - >>> 'epoch': 10, - >>> 'rank1': 0.5, - >>> 'optimizer': optimizer.state_dict() - >>> } - >>> save_checkpoint(state, 'log/my_model') - """ - mkdir_if_missing(save_dir) - if remove_module_from_keys: - # remove 'module.' in state_dict's keys - state_dict = state['state_dict'] - new_state_dict = OrderedDict() - for k, v in state_dict.items(): - if k.startswith('module.'): - k = k[7:] - new_state_dict[k] = v - state['state_dict'] = new_state_dict - # save - epoch = state['epoch'] - fpath = osp.join(save_dir, 'model.pth.tar-' + str(epoch)) - torch.save(state, fpath) - print('Checkpoint saved to "{}"'.format(fpath)) - if is_best: - shutil.copy(fpath, osp.join(osp.dirname(fpath), 'model-best.pth.tar')) - - -def load_checkpoint(fpath): - r"""Loads checkpoint. - - ``UnicodeDecodeError`` can be well handled, which means - python2-saved files can be read from python3. - - Args: - fpath (str): path to checkpoint. - - Returns: - dict - - Examples:: - >>> from torchreid.utils import load_checkpoint - >>> fpath = 'log/my_model/model.pth.tar-10' - >>> checkpoint = load_checkpoint(fpath) - """ - if fpath is None: - raise ValueError('File path is None') - if not osp.exists(fpath): - raise FileNotFoundError('File is not found at "{}"'.format(fpath)) - map_location = None if torch.cuda.is_available() else 'cpu' - try: - checkpoint = torch.load(fpath, map_location=map_location) - except UnicodeDecodeError: - pickle.load = partial(pickle.load, encoding="latin1") - pickle.Unpickler = partial(pickle.Unpickler, encoding="latin1") - checkpoint = torch.load( - fpath, pickle_module=pickle, map_location=map_location - ) - except Exception: - print('Unable to load checkpoint from "{}"'.format(fpath)) - raise - return checkpoint - - -def resume_from_checkpoint(fpath, model, optimizer=None, scheduler=None): - r"""Resumes training from a checkpoint. - - This will load (1) model weights and (2) ``state_dict`` - of optimizer if ``optimizer`` is not None. - - Args: - fpath (str): path to checkpoint. - model (nn.Module): model. - optimizer (Optimizer, optional): an Optimizer. - scheduler (LRScheduler, optional): an LRScheduler. - - Returns: - int: start_epoch. - - Examples:: - >>> from torchreid.utils import resume_from_checkpoint - >>> fpath = 'log/my_model/model.pth.tar-10' - >>> start_epoch = resume_from_checkpoint( - >>> fpath, model, optimizer, scheduler - >>> ) - """ - print('Loading checkpoint from "{}"'.format(fpath)) - checkpoint = load_checkpoint(fpath) - model.load_state_dict(checkpoint['state_dict']) - print('Loaded model weights') - if optimizer is not None and 'optimizer' in checkpoint.keys(): - optimizer.load_state_dict(checkpoint['optimizer']) - print('Loaded optimizer') - if scheduler is not None and 'scheduler' in checkpoint.keys(): - scheduler.load_state_dict(checkpoint['scheduler']) - print('Loaded scheduler') - start_epoch = checkpoint['epoch'] - print('Last epoch = {}'.format(start_epoch)) - if 'rank1' in checkpoint.keys(): - print('Last rank1 = {:.1%}'.format(checkpoint['rank1'])) - return start_epoch - - -def adjust_learning_rate( - optimizer, - base_lr, - epoch, - stepsize=20, - gamma=0.1, - linear_decay=False, - final_lr=0, - max_epoch=100 -): - r"""Adjusts learning rate. - - Deprecated. - """ - if linear_decay: - # linearly decay learning rate from base_lr to final_lr - frac_done = epoch / max_epoch - lr = frac_done*final_lr + (1.-frac_done) * base_lr - else: - # decay learning rate by gamma for every stepsize - lr = base_lr * (gamma**(epoch // stepsize)) - - for param_group in optimizer.param_groups: - param_group['lr'] = lr - - -def set_bn_to_eval(m): - r"""Sets BatchNorm layers to eval mode.""" - # 1. no update for running mean and var - # 2. scale and shift parameters are still trainable - classname = m.__class__.__name__ - if classname.find('BatchNorm') != -1: - m.eval() - - -def open_all_layers(model): - r"""Opens all layers in model for training. - - Examples:: - >>> from torchreid.utils import open_all_layers - >>> open_all_layers(model) - """ - model.train() - for p in model.parameters(): - p.requires_grad = True - - -def open_specified_layers(model, open_layers): - r"""Opens specified layers in model for training while keeping - other layers frozen. - - Args: - model (nn.Module): neural net model. - open_layers (str or list): layers open for training. - - Examples:: - >>> from torchreid.utils import open_specified_layers - >>> # Only model.classifier will be updated. - >>> open_layers = 'classifier' - >>> open_specified_layers(model, open_layers) - >>> # Only model.fc and model.classifier will be updated. - >>> open_layers = ['fc', 'classifier'] - >>> open_specified_layers(model, open_layers) - """ - if isinstance(model, nn.DataParallel): - model = model.module - elif isinstance(model, nn.parallel.DistributedDataParallel): - model = model.module - - if isinstance(open_layers, str): - open_layers = [open_layers] - - for layer in open_layers: - assert hasattr( - model, layer - ), '"{}" is not an attribute of the model, please provide the correct name'.format( - layer - ) - - for name, module in model.named_children(): - if name in open_layers: - module.train() - for p in module.parameters(): - p.requires_grad = True - else: - module.eval() - for p in module.parameters(): - p.requires_grad = False - - -def count_num_param(model): - r"""Counts number of parameters in a model while ignoring ``self.classifier``. - - Args: - model (nn.Module): network model. - - Examples:: - >>> from torchreid.utils import count_num_param - >>> model_size = count_num_param(model) - - .. warning:: - - This method is deprecated in favor of - ``torchreid.utils.compute_model_complexity``. - """ - warnings.warn( - 'This method is deprecated and will be removed in the future.' - ) - - num_param = sum(p.numel() for p in model.parameters()) - - if isinstance(model, nn.DataParallel): - model = model.module - - if hasattr(model, - 'classifier') and isinstance(model.classifier, nn.Module): - # we ignore the classifier because it is unused at test time - num_param -= sum(p.numel() for p in model.classifier.parameters()) - - return num_param - - -def load_pretrained_weights(model, weight_path, ignore_classifier=False): - r"""Loads pretrianed weights to model. - - Features:: - - Incompatible layers (unmatched in name or size) will be ignored. - - Can automatically deal with keys containing "module.". - - Args: - model (nn.Module): network model. - weight_path (str): path to pretrained weights. - - Examples:: - >>> from torchreid.utils import load_pretrained_weights - >>> weight_path = 'log/my_model/model-best.pth.tar' - >>> load_pretrained_weights(model, weight_path) - """ - checkpoint = load_checkpoint(weight_path) - if 'state_dict' in checkpoint: - state_dict = checkpoint['state_dict'] - else: - state_dict = checkpoint - - model_dict = model.state_dict() - new_state_dict = OrderedDict() - matched_layers, discarded_layers = [], [] - - for k, v in state_dict.items(): - if k.startswith('module.'): - k = k[7:] # discard module. - - if ignore_classifier and k.startswith('classifier'): - continue - - if k in model_dict and model_dict[k].size() == v.size(): - new_state_dict[k] = v - matched_layers.append(k) - else: - discarded_layers.append(k) - - model_dict.update(new_state_dict) - model.load_state_dict(model_dict) - - if len(matched_layers) == 0: - warnings.warn( - 'The pretrained weights "{}" cannot be loaded, ' - 'please check the key names manually ' - '(** ignored and continue **)'.format(weight_path) - ) - else: - print( - 'Successfully loaded pretrained weights from "{}"'. - format(weight_path) - ) - if len(discarded_layers) > 0: - print( - '** The following layers are discarded ' - 'due to unmatched keys or layer size: {}'. - format(discarded_layers) - ) -- Gitee From 1615642d2ecf29a53c55ebe5af0dee8eeb7c5e9f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:51:18 +0000 Subject: [PATCH 10/31] init --- .../OSNet/torchreid/__init__.py | 56 ++ .../OSNet/torchreid/data/__init__.py | 54 ++ .../OSNet/torchreid/data/datamanager.py | 609 ++++++++++++++++++ .../OSNet/torchreid/data/sampler.py | 292 +++++++++ .../OSNet/torchreid/data/transforms.py | 373 +++++++++++ .../OSNet/torchreid/engine/__init__.py | 52 ++ .../OSNet/torchreid/engine/engine.py | 547 ++++++++++++++++ .../OSNet/torchreid/losses/__init__.py | 68 ++ .../torchreid/losses/cross_entropy_loss.py | 100 +++ .../losses/hard_mine_triplet_loss.py | 95 +++ .../OSNet/torchreid/metrics/__init__.py | 52 ++ .../OSNet/torchreid/metrics/accuracy.py | 84 +++ .../OSNet/torchreid/metrics/distance.py | 127 ++++ .../OSNet/torchreid/metrics/rank.py | 254 ++++++++ .../OSNet/torchreid/models/densenet.py | 425 ++++++++++++ .../OSNet/torchreid/models/hacnn.py | 461 +++++++++++++ .../torchreid/models/inceptionresnetv2.py | 406 ++++++++++++ .../OSNet/torchreid/models/inceptionv4.py | 428 ++++++++++++ .../OSNet/torchreid/models/mlfn.py | 316 +++++++++ .../OSNet/torchreid/models/mobilenetv2.py | 321 +++++++++ 20 files changed, 5120 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py new file mode 100644 index 0000000000..40669b944c --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py @@ -0,0 +1,56 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from torchreid import data, optim, utils, engine, losses, models, metrics + +__version__ = '1.4.0' +__author__ = 'Kaiyang Zhou' +__homepage__ = 'https://kaiyangzhou.github.io/' +__description__ = 'Deep learning person re-identification in PyTorch' +__url__ = 'https://github.com/KaiyangZhou/deep-person-reid' diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py new file mode 100644 index 0000000000..577ed45750 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py @@ -0,0 +1,54 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .datasets import ( + Dataset, ImageDataset, VideoDataset, register_image_dataset, + register_video_dataset +) +from .datamanager import ImageDataManager, VideoDataManager diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py new file mode 100644 index 0000000000..009d03802f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py @@ -0,0 +1,609 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch + +from torchreid.data.sampler import build_train_sampler +from torchreid.data.datasets import init_image_dataset, init_video_dataset +from torchreid.data.transforms import build_transforms + + +class DataManager(object): + r"""Base data manager. + + Args: + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + """ + + def __init__( + self, + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + norm_mean=None, + norm_std=None, + use_gpu=False + ): + self.sources = sources + self.targets = targets + self.height = height + self.width = width + + if self.sources is None: + raise ValueError('sources must not be None') + + if isinstance(self.sources, str): + self.sources = [self.sources] + + if self.targets is None: + self.targets = self.sources + + if isinstance(self.targets, str): + self.targets = [self.targets] + + self.transform_tr, self.transform_te = build_transforms( + self.height, + self.width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std + ) + + self.use_gpu = (torch.cuda.is_available() and use_gpu) + + @property + def num_train_pids(self): + """Returns the number of training person identities.""" + return self._num_train_pids + + @property + def num_train_cams(self): + """Returns the number of training cameras.""" + return self._num_train_cams + + def fetch_test_loaders(self, name): + """Returns query and gallery of a test dataset, each containing + tuples of (img_path(s), pid, camid). + + Args: + name (str): dataset name. + """ + query_loader = self.test_dataset[name]['query'] + gallery_loader = self.test_dataset[name]['gallery'] + return query_loader, gallery_loader + + def preprocess_pil_img(self, img): + """Transforms a PIL image to torch tensor for testing.""" + return self.transform_te(img) + + +class ImageDataManager(DataManager): + r"""Image data manager. + + Args: + root (str): root path to datasets. + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + k_tfm (int): number of times to apply augmentation to an image + independently. If k_tfm > 1, the transform function will be + applied k_tfm times to an image. This variable will only be + useful for training and is currently valid for image datasets only. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + split_id (int, optional): split id (*0-based*). Default is 0. + combineall (bool, optional): combine train, query and gallery in a dataset for + training. Default is False. + load_train_targets (bool, optional): construct train-loader for target datasets. + Default is False. This is useful for domain adaptation research. + batch_size_train (int, optional): number of images in a training batch. Default is 32. + batch_size_test (int, optional): number of images in a test batch. Default is 32. + workers (int, optional): number of workers. Default is 4. + num_instances (int, optional): number of instances per identity in a batch. + Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + train_sampler (str, optional): sampler. Default is RandomSampler. + train_sampler_t (str, optional): sampler for target train loader. Default is RandomSampler. + cuhk03_labeled (bool, optional): use cuhk03 labeled images. + Default is False (defaul is to use detected images). + cuhk03_classic_split (bool, optional): use the classic split in cuhk03. + Default is False. + market1501_500k (bool, optional): add 500K distractors to the gallery + set in market1501. Default is False. + + Examples:: + + datamanager = torchreid.data.ImageDataManager( + root='path/to/reid-data', + sources='market1501', + height=256, + width=128, + batch_size_train=32, + batch_size_test=100 + ) + + # return train loader of source data + train_loader = datamanager.train_loader + + # return test loader of target data + test_loader = datamanager.test_loader + + # return train loader of target data + train_loader_t = datamanager.train_loader_t + """ + data_type = 'image' + + def __init__( + self, + root='', + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + k_tfm=1, + norm_mean=None, + norm_std=None, + use_gpu=True, + split_id=0, + combineall=False, + load_train_targets=False, + batch_size_train=32, + batch_size_test=32, + workers=4, + num_instances=4, + num_cams=1, + num_datasets=1, + train_sampler='RandomSampler', + train_sampler_t='RandomSampler', + cuhk03_labeled=False, + cuhk03_classic_split=False, + market1501_500k=False, + device_num=-1 + ): + + super(ImageDataManager, self).__init__( + sources=sources, + targets=targets, + height=height, + width=width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std, + use_gpu=use_gpu + ) + + print('=> Loading train (source) dataset') + trainset = [] + for name in self.sources: + trainset_ = init_image_dataset( + name, + transform=self.transform_tr, + k_tfm=k_tfm, + mode='train', + combineall=combineall, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + trainset.append(trainset_) + trainset = sum(trainset) + + self._num_train_pids = trainset.num_train_pids + self._num_train_cams = trainset.num_train_cams + + if device_num == -1 or device_num == 1: + self.train_sampler = build_train_sampler( + trainset.train, + train_sampler, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ) + + else: + self.train_sampler = torch.utils.data.distributed.DistributedSampler(trainset.train) + + self.train_loader = torch.utils.data.DataLoader( + trainset, + sampler=self.train_sampler, + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + # pin_memory=self.use_gpu, + pin_memory=True, + drop_last=True + ) + + self.train_loader_t = None + if load_train_targets: + # check if sources and targets are identical + assert len(set(self.sources) & set(self.targets)) == 0, \ + 'sources={} and targets={} must not have overlap'.format(self.sources, self.targets) + + print('=> Loading train (target) dataset') + trainset_t = [] + for name in self.targets: + trainset_t_ = init_image_dataset( + name, + transform=self.transform_tr, + k_tfm=k_tfm, + mode='train', + combineall=False, # only use the training data + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + trainset_t.append(trainset_t_) + trainset_t = sum(trainset_t) + + self.train_loader_t = torch.utils.data.DataLoader( + trainset_t, + sampler=build_train_sampler( + trainset_t.train, + train_sampler_t, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ), + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=True + ) + + print('=> Loading test (target) dataset') + self.test_loader = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + self.test_dataset = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + + for name in self.targets: + # build query loader + queryset = init_image_dataset( + name, + transform=self.transform_te, + mode='query', + combineall=combineall, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + self.test_loader[name]['query'] = torch.utils.data.DataLoader( + queryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=True, + drop_last=False + ) + + # build gallery loader + galleryset = init_image_dataset( + name, + transform=self.transform_te, + mode='gallery', + combineall=combineall, + verbose=False, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( + galleryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=True, + drop_last=False + ) + + self.test_dataset[name]['query'] = queryset.query + self.test_dataset[name]['gallery'] = galleryset.gallery + + print('\n') + print(' **************** Summary ****************') + print(' source : {}'.format(self.sources)) + print(' # source datasets : {}'.format(len(self.sources))) + print(' # source ids : {}'.format(self.num_train_pids)) + print(' # source images : {}'.format(len(trainset))) + print(' # source cameras : {}'.format(self.num_train_cams)) + if load_train_targets: + print( + ' # target images : {} (unlabeled)'.format(len(trainset_t)) + ) + print(' target : {}'.format(self.targets)) + print(' *****************************************') + print('\n') + + +class VideoDataManager(DataManager): + r"""Video data manager. + + Args: + root (str): root path to datasets. + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + split_id (int, optional): split id (*0-based*). Default is 0. + combineall (bool, optional): combine train, query and gallery in a dataset for + training. Default is False. + batch_size_train (int, optional): number of tracklets in a training batch. Default is 3. + batch_size_test (int, optional): number of tracklets in a test batch. Default is 3. + workers (int, optional): number of workers. Default is 4. + num_instances (int, optional): number of instances per identity in a batch. + Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + train_sampler (str, optional): sampler. Default is RandomSampler. + seq_len (int, optional): how many images to sample in a tracklet. Default is 15. + sample_method (str, optional): how to sample images in a tracklet. Default is "evenly". + Choices are ["evenly", "random", "all"]. "evenly" and "random" will sample ``seq_len`` + images in a tracklet while "all" samples all images in a tracklet, where the batch size + needs to be set to 1. + + Examples:: + + datamanager = torchreid.data.VideoDataManager( + root='path/to/reid-data', + sources='mars', + height=256, + width=128, + batch_size_train=3, + batch_size_test=3, + seq_len=15, + sample_method='evenly' + ) + + # return train loader of source data + train_loader = datamanager.train_loader + + # return test loader of target data + test_loader = datamanager.test_loader + + .. note:: + The current implementation only supports image-like training. Therefore, each image in a + sampled tracklet will undergo independent transformation functions. To achieve tracklet-aware + training, you need to modify the transformation functions for video reid such that each function + applies the same operation to all images in a tracklet to keep consistency. + """ + data_type = 'video' + + def __init__( + self, + root='', + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + norm_mean=None, + norm_std=None, + use_gpu=True, + split_id=0, + combineall=False, + batch_size_train=3, + batch_size_test=3, + workers=4, + num_instances=4, + num_cams=1, + num_datasets=1, + train_sampler='RandomSampler', + seq_len=15, + sample_method='evenly' + ): + + super(VideoDataManager, self).__init__( + sources=sources, + targets=targets, + height=height, + width=width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std, + use_gpu=use_gpu + ) + + print('=> Loading train (source) dataset') + trainset = [] + for name in self.sources: + trainset_ = init_video_dataset( + name, + transform=self.transform_tr, + mode='train', + combineall=combineall, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + trainset.append(trainset_) + trainset = sum(trainset) + + self._num_train_pids = trainset.num_train_pids + self._num_train_cams = trainset.num_train_cams + + train_sampler = build_train_sampler( + trainset.train, + train_sampler, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ) + + self.train_loader = torch.utils.data.DataLoader( + trainset, + sampler=train_sampler, + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=True + ) + + print('=> Loading test (target) dataset') + self.test_loader = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + self.test_dataset = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + + for name in self.targets: + # build query loader + queryset = init_video_dataset( + name, + transform=self.transform_te, + mode='query', + combineall=combineall, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + self.test_loader[name]['query'] = torch.utils.data.DataLoader( + queryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=False + ) + + # build gallery loader + galleryset = init_video_dataset( + name, + transform=self.transform_te, + mode='gallery', + combineall=combineall, + verbose=False, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( + galleryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=False + ) + + self.test_dataset[name]['query'] = queryset.query + self.test_dataset[name]['gallery'] = galleryset.gallery + + print('\n') + print(' **************** Summary ****************') + print(' source : {}'.format(self.sources)) + print(' # source datasets : {}'.format(len(self.sources))) + print(' # source ids : {}'.format(self.num_train_pids)) + print(' # source tracklets : {}'.format(len(trainset))) + print(' # source cameras : {}'.format(self.num_train_cams)) + print(' target : {}'.format(self.targets)) + print(' *****************************************') + print('\n') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py new file mode 100644 index 0000000000..daf0d026c3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py @@ -0,0 +1,292 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import copy +import numpy as np +import random +from collections import defaultdict +from torch.utils.data.sampler import Sampler, RandomSampler, SequentialSampler + +AVAI_SAMPLERS = [ + 'RandomIdentitySampler', 'SequentialSampler', 'RandomSampler', + 'RandomDomainSampler', 'RandomDatasetSampler' +] + + +class RandomIdentitySampler(Sampler): + """Randomly samples N identities each with K instances. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + num_instances (int): number of instances per identity in a batch. + """ + + def __init__(self, data_source, batch_size, num_instances): + if batch_size < num_instances: + raise ValueError( + 'batch_size={} must be no less ' + 'than num_instances={}'.format(batch_size, num_instances) + ) + + self.data_source = data_source + self.batch_size = batch_size + self.num_instances = num_instances + self.num_pids_per_batch = self.batch_size // self.num_instances + self.index_dic = defaultdict(list) + for index, items in enumerate(data_source): + pid = items[1] + self.index_dic[pid].append(index) + self.pids = list(self.index_dic.keys()) + assert len(self.pids) >= self.num_pids_per_batch + + # estimate number of examples in an epoch + # TODO: improve precision + self.length = 0 + for pid in self.pids: + idxs = self.index_dic[pid] + num = len(idxs) + if num < self.num_instances: + num = self.num_instances + self.length += num - num % self.num_instances + + def __iter__(self): + batch_idxs_dict = defaultdict(list) + + for pid in self.pids: + idxs = copy.deepcopy(self.index_dic[pid]) + if len(idxs) < self.num_instances: + idxs = np.random.choice( + idxs, size=self.num_instances, replace=True + ) + random.shuffle(idxs) + batch_idxs = [] + for idx in idxs: + batch_idxs.append(idx) + if len(batch_idxs) == self.num_instances: + batch_idxs_dict[pid].append(batch_idxs) + batch_idxs = [] + + avai_pids = copy.deepcopy(self.pids) + final_idxs = [] + + while len(avai_pids) >= self.num_pids_per_batch: + selected_pids = random.sample(avai_pids, self.num_pids_per_batch) + for pid in selected_pids: + batch_idxs = batch_idxs_dict[pid].pop(0) + final_idxs.extend(batch_idxs) + if len(batch_idxs_dict[pid]) == 0: + avai_pids.remove(pid) + + return iter(final_idxs) + + def __len__(self): + return self.length + + +class RandomDomainSampler(Sampler): + """Random domain sampler. + + We consider each camera as a visual domain. + + How does the sampling work: + 1. Randomly sample N cameras (based on the "camid" label). + 2. From each camera, randomly sample K images. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + n_domain (int): number of cameras to sample in a batch. + """ + + def __init__(self, data_source, batch_size, n_domain): + self.data_source = data_source + + # Keep track of image indices for each domain + self.domain_dict = defaultdict(list) + for i, items in enumerate(data_source): + camid = items[2] + self.domain_dict[camid].append(i) + self.domains = list(self.domain_dict.keys()) + + # Make sure each domain can be assigned an equal number of images + if n_domain is None or n_domain <= 0: + n_domain = len(self.domains) + assert batch_size % n_domain == 0 + self.n_img_per_domain = batch_size // n_domain + + self.batch_size = batch_size + self.n_domain = n_domain + self.length = len(list(self.__iter__())) + + def __iter__(self): + domain_dict = copy.deepcopy(self.domain_dict) + final_idxs = [] + stop_sampling = False + + while not stop_sampling: + selected_domains = random.sample(self.domains, self.n_domain) + + for domain in selected_domains: + idxs = domain_dict[domain] + selected_idxs = random.sample(idxs, self.n_img_per_domain) + final_idxs.extend(selected_idxs) + + for idx in selected_idxs: + domain_dict[domain].remove(idx) + + remaining = len(domain_dict[domain]) + if remaining < self.n_img_per_domain: + stop_sampling = True + + return iter(final_idxs) + + def __len__(self): + return self.length + + +class RandomDatasetSampler(Sampler): + """Random dataset sampler. + + How does the sampling work: + 1. Randomly sample N datasets (based on the "dsetid" label). + 2. From each dataset, randomly sample K images. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + n_dataset (int): number of datasets to sample in a batch. + """ + + def __init__(self, data_source, batch_size, n_dataset): + self.data_source = data_source + + # Keep track of image indices for each dataset + self.dataset_dict = defaultdict(list) + for i, items in enumerate(data_source): + dsetid = items[3] + self.dataset_dict[dsetid].append(i) + self.datasets = list(self.dataset_dict.keys()) + + # Make sure each dataset can be assigned an equal number of images + if n_dataset is None or n_dataset <= 0: + n_dataset = len(self.datasets) + assert batch_size % n_dataset == 0 + self.n_img_per_dset = batch_size // n_dataset + + self.batch_size = batch_size + self.n_dataset = n_dataset + self.length = len(list(self.__iter__())) + + def __iter__(self): + dataset_dict = copy.deepcopy(self.dataset_dict) + final_idxs = [] + stop_sampling = False + + while not stop_sampling: + selected_datasets = random.sample(self.datasets, self.n_dataset) + + for dset in selected_datasets: + idxs = dataset_dict[dset] + selected_idxs = random.sample(idxs, self.n_img_per_dset) + final_idxs.extend(selected_idxs) + + for idx in selected_idxs: + dataset_dict[dset].remove(idx) + + remaining = len(dataset_dict[dset]) + if remaining < self.n_img_per_dset: + stop_sampling = True + + return iter(final_idxs) + + def __len__(self): + return self.length + + +def build_train_sampler( + data_source, + train_sampler, + batch_size=32, + num_instances=4, + num_cams=1, + num_datasets=1, + **kwargs +): + """Builds a training sampler. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid). + train_sampler (str): sampler name (default: ``RandomSampler``). + batch_size (int, optional): batch size. Default is 32. + num_instances (int, optional): number of instances per identity in a + batch (when using ``RandomIdentitySampler``). Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + """ + assert train_sampler in AVAI_SAMPLERS, \ + 'train_sampler must be one of {}, but got {}'.format(AVAI_SAMPLERS, train_sampler) + + if train_sampler == 'RandomIdentitySampler': + sampler = RandomIdentitySampler(data_source, batch_size, num_instances) + + elif train_sampler == 'RandomDomainSampler': + sampler = RandomDomainSampler(data_source, batch_size, num_cams) + + elif train_sampler == 'RandomDatasetSampler': + sampler = RandomDatasetSampler(data_source, batch_size, num_datasets) + + elif train_sampler == 'SequentialSampler': + sampler = SequentialSampler(data_source) + + elif train_sampler == 'RandomSampler': + sampler = RandomSampler(data_source) + + return sampler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py new file mode 100644 index 0000000000..3108b81565 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py @@ -0,0 +1,373 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import math +import random +from collections import deque +import torch +from PIL import Image +from torchvision.transforms import ( + Resize, Compose, ToTensor, Normalize, ColorJitter, RandomHorizontalFlip +) + + +class Random2DTranslation(object): + """Randomly translates the input image with a probability. + + Specifically, given a predefined shape (height, width), the input is first + resized with a factor of 1.125, leading to (height*1.125, width*1.125), then + a random crop is performed. Such operation is done with a probability. + + Args: + height (int): target image height. + width (int): target image width. + p (float, optional): probability that this operation takes place. + Default is 0.5. + interpolation (int, optional): desired interpolation. Default is + ``PIL.Image.BILINEAR`` + """ + + def __init__(self, height, width, p=0.5, interpolation=Image.BILINEAR): + self.height = height + self.width = width + self.p = p + self.interpolation = interpolation + + def __call__(self, img): + if random.uniform(0, 1) > self.p: + return img.resize((self.width, self.height), self.interpolation) + + new_width, new_height = int(round(self.width * 1.125) + ), int(round(self.height * 1.125)) + resized_img = img.resize((new_width, new_height), self.interpolation) + x_maxrange = new_width - self.width + y_maxrange = new_height - self.height + x1 = int(round(random.uniform(0, x_maxrange))) + y1 = int(round(random.uniform(0, y_maxrange))) + croped_img = resized_img.crop( + (x1, y1, x1 + self.width, y1 + self.height) + ) + return croped_img + + +class RandomErasing(object): + """Randomly erases an image patch. + + Origin: ``_ + + Reference: + Zhong et al. Random Erasing Data Augmentation. + + Args: + probability (float, optional): probability that this operation takes place. + Default is 0.5. + sl (float, optional): min erasing area. + sh (float, optional): max erasing area. + r1 (float, optional): min aspect ratio. + mean (list, optional): erasing value. + """ + + def __init__( + self, + probability=0.5, + sl=0.02, + sh=0.4, + r1=0.3, + mean=[0.4914, 0.4822, 0.4465] + ): + self.probability = probability + self.mean = mean + self.sl = sl + self.sh = sh + self.r1 = r1 + + def __call__(self, img): + if random.uniform(0, 1) > self.probability: + return img + + for attempt in range(100): + area = img.size()[1] * img.size()[2] + + target_area = random.uniform(self.sl, self.sh) * area + aspect_ratio = random.uniform(self.r1, 1 / self.r1) + + h = int(round(math.sqrt(target_area * aspect_ratio))) + w = int(round(math.sqrt(target_area / aspect_ratio))) + + if w < img.size()[2] and h < img.size()[1]: + x1 = random.randint(0, img.size()[1] - h) + y1 = random.randint(0, img.size()[2] - w) + if img.size()[0] == 3: + img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] + img[1, x1:x1 + h, y1:y1 + w] = self.mean[1] + img[2, x1:x1 + h, y1:y1 + w] = self.mean[2] + else: + img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] + return img + + return img + + +class ColorAugmentation(object): + """Randomly alters the intensities of RGB channels. + + Reference: + Krizhevsky et al. ImageNet Classification with Deep ConvolutionalNeural + Networks. NIPS 2012. + + Args: + p (float, optional): probability that this operation takes place. + Default is 0.5. + """ + + def __init__(self, p=0.5): + self.p = p + self.eig_vec = torch.Tensor( + [ + [0.4009, 0.7192, -0.5675], + [-0.8140, -0.0045, -0.5808], + [0.4203, -0.6948, -0.5836], + ] + ) + self.eig_val = torch.Tensor([[0.2175, 0.0188, 0.0045]]) + + def _check_input(self, tensor): + assert tensor.dim() == 3 and tensor.size(0) == 3 + + def __call__(self, tensor): + if random.uniform(0, 1) > self.p: + return tensor + alpha = torch.normal(mean=torch.zeros_like(self.eig_val)) * 0.1 + quatity = torch.mm(self.eig_val * alpha, self.eig_vec) + tensor = tensor + quatity.view(3, 1, 1) + return tensor + + +class RandomPatch(object): + """Random patch data augmentation. + + There is a patch pool that stores randomly extracted pathces from person images. + + For each input image, RandomPatch + 1) extracts a random patch and stores the patch in the patch pool; + 2) randomly selects a patch from the patch pool and pastes it on the + input (at random position) to simulate occlusion. + + Reference: + - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. + - Zhou et al. Learning Generalisable Omni-Scale Representations + for Person Re-Identification. TPAMI, 2021. + """ + + def __init__( + self, + prob_happen=0.5, + pool_capacity=50000, + min_sample_size=100, + patch_min_area=0.01, + patch_max_area=0.5, + patch_min_ratio=0.1, + prob_rotate=0.5, + prob_flip_leftright=0.5, + ): + self.prob_happen = prob_happen + + self.patch_min_area = patch_min_area + self.patch_max_area = patch_max_area + self.patch_min_ratio = patch_min_ratio + + self.prob_rotate = prob_rotate + self.prob_flip_leftright = prob_flip_leftright + + self.patchpool = deque(maxlen=pool_capacity) + self.min_sample_size = min_sample_size + + def generate_wh(self, W, H): + area = W * H + for attempt in range(100): + target_area = random.uniform( + self.patch_min_area, self.patch_max_area + ) * area + aspect_ratio = random.uniform( + self.patch_min_ratio, 1. / self.patch_min_ratio + ) + h = int(round(math.sqrt(target_area * aspect_ratio))) + w = int(round(math.sqrt(target_area / aspect_ratio))) + if w < W and h < H: + return w, h + return None, None + + def transform_patch(self, patch): + if random.uniform(0, 1) > self.prob_flip_leftright: + patch = patch.transpose(Image.FLIP_LEFT_RIGHT) + if random.uniform(0, 1) > self.prob_rotate: + patch = patch.rotate(random.randint(-10, 10)) + return patch + + def __call__(self, img): + W, H = img.size # original image size + + # collect new patch + w, h = self.generate_wh(W, H) + if w is not None and h is not None: + x1 = random.randint(0, W - w) + y1 = random.randint(0, H - h) + new_patch = img.crop((x1, y1, x1 + w, y1 + h)) + self.patchpool.append(new_patch) + + if len(self.patchpool) < self.min_sample_size: + return img + + if random.uniform(0, 1) > self.prob_happen: + return img + + # paste a randomly selected patch on a random position + patch = random.sample(self.patchpool, 1)[0] + patchW, patchH = patch.size + x1 = random.randint(0, W - patchW) + y1 = random.randint(0, H - patchH) + patch = self.transform_patch(patch) + img.paste(patch, (x1, y1)) + + return img + + +def build_transforms( + height, + width, + transforms='random_flip', + norm_mean=[0.485, 0.456, 0.406], + norm_std=[0.229, 0.224, 0.225], + **kwargs +): + """Builds train and test transform functions. + + Args: + height (int): target image height. + width (int): target image width. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): normalization mean values. Default is ImageNet means. + norm_std (list or None, optional): normalization standard deviation values. Default is + ImageNet standard deviation values. + """ + if transforms is None: + transforms = [] + + if isinstance(transforms, str): + transforms = [transforms] + + if not isinstance(transforms, list): + raise ValueError( + 'transforms must be a list of strings, but found to be {}'.format( + type(transforms) + ) + ) + + if len(transforms) > 0: + transforms = [t.lower() for t in transforms] + + if norm_mean is None or norm_std is None: + norm_mean = [0.485, 0.456, 0.406] # imagenet mean + norm_std = [0.229, 0.224, 0.225] # imagenet std + normalize = Normalize(mean=norm_mean, std=norm_std) + + print('Building train transforms ...') + transform_tr = [] + + print('+ resize to {}x{}'.format(height, width)) + transform_tr += [Resize((height, width))] + + if 'random_flip' in transforms: + print('+ random flip') + transform_tr += [RandomHorizontalFlip()] + + if 'random_crop' in transforms: + print( + '+ random crop (enlarge to {}x{} and ' + 'crop {}x{})'.format( + int(round(height * 1.125)), int(round(width * 1.125)), height, + width + ) + ) + transform_tr += [Random2DTranslation(height, width)] + + if 'random_patch' in transforms: + print('+ random patch') + transform_tr += [RandomPatch()] + + if 'color_jitter' in transforms: + print('+ color jitter') + transform_tr += [ + ColorJitter(brightness=0.2, contrast=0.15, saturation=0, hue=0) + ] + + print('+ to torch tensor of range [0, 1]') + transform_tr += [ToTensor()] + + print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) + transform_tr += [normalize] + + if 'random_erase' in transforms: + print('+ random erase') + transform_tr += [RandomErasing(mean=norm_mean)] + + transform_tr = Compose(transform_tr) + + print('Building test transforms ...') + print('+ resize to {}x{}'.format(height, width)) + print('+ to torch tensor of range [0, 1]') + print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) + + transform_te = Compose([ + Resize((height, width)), + ToTensor(), + normalize, + ]) + + return transform_tr, transform_te diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py new file mode 100644 index 0000000000..7eca9586f7 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py @@ -0,0 +1,52 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .image import ImageSoftmaxEngine, ImageTripletEngine +from .video import VideoSoftmaxEngine, VideoTripletEngine +from .engine import Engine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py new file mode 100644 index 0000000000..4dd250fa95 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py @@ -0,0 +1,547 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import time +import numpy as np +import os.path as osp +import datetime +from collections import OrderedDict +import torch +from torch.nn import functional as F +from torch.utils.tensorboard import SummaryWriter + +from torchreid import metrics +from torchreid.utils import ( + MetricMeter, AverageMeter, re_ranking, open_all_layers, save_checkpoint, + open_specified_layers, visualize_ranked_results +) +from torchreid.losses import DeepSupervision +import os + + +class Engine(object): + r"""A generic base Engine class for both image- and video-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + use_gpu (bool, optional): use gpu. Default is True. + """ + + def __init__(self, datamanager, use_gpu=False, use_npu=False): + self.datamanager = datamanager + self.train_loader = self.datamanager.train_loader + self.test_loader = self.datamanager.test_loader + self.use_gpu = use_gpu + self.use_npu = use_npu + self.writer = None + self.epoch = 0 + + self.model = None + self.optimizer = None + self.scheduler = None + + self._models = OrderedDict() + self._optims = OrderedDict() + self._scheds = OrderedDict() + + def register_model(self, name='model', model=None, optim=None, sched=None): + if self.__dict__.get('_models') is None: + raise AttributeError( + 'Cannot assign model before super().__init__() call' + ) + + if self.__dict__.get('_optims') is None: + raise AttributeError( + 'Cannot assign optim before super().__init__() call' + ) + + if self.__dict__.get('_scheds') is None: + raise AttributeError( + 'Cannot assign sched before super().__init__() call' + ) + + self._models[name] = model + self._optims[name] = optim + self._scheds[name] = sched + + def get_model_names(self, names=None): + names_real = list(self._models.keys()) + if names is not None: + if not isinstance(names, list): + names = [names] + for name in names: + assert name in names_real + return names + else: + return names_real + + def save_model(self, epoch, rank1, save_dir, is_best=False): + names = self.get_model_names() + + for name in names: + save_checkpoint( + { + 'state_dict': self._models[name].state_dict(), + 'epoch': epoch + 1, + 'rank1': rank1, + 'optimizer': self._optims[name].state_dict(), + 'scheduler': self._scheds[name].state_dict() + }, + osp.join(save_dir, name), + is_best=is_best + ) + + def set_model_mode(self, mode='train', names=None): + assert mode in ['train', 'eval', 'test'] + names = self.get_model_names(names) + + for name in names: + if mode == 'train': + self._models[name].train() + else: + self._models[name].eval() + + def get_current_lr(self, names=None): + names = self.get_model_names(names) + name = names[0] + return self._optims[name].param_groups[-1]['lr'] + + def update_lr(self, names=None): + names = self.get_model_names(names) + + for name in names: + if self._scheds[name] is not None: + self._scheds[name].step() + + def run( + self, + save_dir='log', + max_epoch=0, + start_epoch=0, + print_freq=10, + fixbase_epoch=0, + open_layers=None, + start_eval=0, + eval_freq=-1, + test_only=False, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + r"""A unified pipeline for training and evaluating a model. + + Args: + save_dir (str): directory to save model. + max_epoch (int): maximum epoch. + start_epoch (int, optional): starting epoch. Default is 0. + print_freq (int, optional): print_frequency. Default is 10. + fixbase_epoch (int, optional): number of epochs to train ``open_layers`` (new layers) + while keeping base layers frozen. Default is 0. ``fixbase_epoch`` is counted + in ``max_epoch``. + open_layers (str or list, optional): layers (attribute names) open for training. + start_eval (int, optional): from which epoch to start evaluation. Default is 0. + eval_freq (int, optional): evaluation frequency. Default is -1 (meaning evaluation + is only performed at the end of training). + test_only (bool, optional): if True, only runs evaluation on test datasets. + Default is False. + dist_metric (str, optional): distance metric used to compute distance matrix + between query and gallery. Default is "euclidean". + normalize_feature (bool, optional): performs L2 normalization on feature vectors before + computing feature distance. Default is False. + visrank (bool, optional): visualizes ranked results. Default is False. It is recommended to + enable ``visrank`` when ``test_only`` is True. The ranked images will be saved to + "save_dir/visrank_dataset", e.g. "save_dir/visrank_market1501". + visrank_topk (int, optional): top-k ranked images to be visualized. Default is 10. + use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. + Default is False. This should be enabled when using cuhk03 classic split. + ranks (list, optional): cmc ranks to be computed. Default is [1, 5, 10, 20]. + rerank (bool, optional): uses person re-ranking (by Zhong et al. CVPR'17). + Default is False. This is only enabled when test_only=True. + """ + + if visrank and not test_only: + raise ValueError( + 'visrank can be set to True only if test_only=True' + ) + + if test_only: + self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks, + rerank=rerank + ) + return + + if self.writer is None: + self.writer = SummaryWriter(log_dir=save_dir) + + time_start = time.time() + self.start_epoch = start_epoch + self.max_epoch = max_epoch + print('=> Start training') + + device_num = int(os.environ['device_num']) + device_num = 1 if device_num == -1 else device_num + batch_size = int(os.environ['batch_size']) + total_avg = 0.0 + + for self.epoch in range(self.start_epoch, self.max_epoch): + if os.environ['device_num'] != '-1' and os.environ['device_num'] != '1': + self.datamanager.train_sampler.set_epoch(self.epoch) + eve_time = self.train( + print_freq=print_freq, + fixbase_epoch=fixbase_epoch, + open_layers=open_layers + ) + total_avg += eve_time + print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / eve_time, eve_time)) + + if (self.epoch + 1) >= start_eval \ + and eval_freq > 0 \ + and (self.epoch+1) % eval_freq == 0 \ + and (self.epoch + 1) != self.max_epoch: + rank1 = self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks + ) + self.save_model(self.epoch, rank1, save_dir) + + avg_time = total_avg / (self.max_epoch - self.start_epoch) + + if self.max_epoch > 1: + print('=> Final test') + rank1 = self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks + ) + self.save_model(self.epoch, rank1, save_dir) + + elapsed = round(time.time() - time_start) + elapsed = str(datetime.timedelta(seconds=elapsed)) + print('Elapsed {}'.format(elapsed)) + + print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / avg_time, avg_time)) + + if self.writer is not None: + self.writer.close() + + def train(self, print_freq=10, fixbase_epoch=0, open_layers=None): + losses = MetricMeter() + batch_time = AverageMeter() + data_time = AverageMeter() + + self.set_model_mode('train') + + self.two_stepped_transfer_learning( + self.epoch, fixbase_epoch, open_layers + ) + + self.num_batches = len(self.train_loader) + end = time.time() + for self.batch_idx, data in enumerate(self.train_loader): + data_time.update(time.time() - end) + loss_summary = self.forward_backward(data) + batch_time.update(time.time() - end) + losses.update(loss_summary) + + if (self.batch_idx + 1) % print_freq == 0: + nb_this_epoch = self.num_batches - (self.batch_idx + 1) + nb_future_epochs = ( + self.max_epoch - (self.epoch + 1) + ) * self.num_batches + eta_seconds = batch_time.avg * (nb_this_epoch+nb_future_epochs) + eta_str = str(datetime.timedelta(seconds=int(eta_seconds))) + print( + 'epoch: [{0}/{1}][{2}/{3}]\t' + 'time {batch_time.val:.3f} ({batch_time.avg:.3f})\t' + 'data {data_time.val:.3f} ({data_time.avg:.3f})\t' + 'eta {eta}\t' + '{losses}\t' + 'lr {lr:.6f}'.format( + self.epoch + 1, + self.max_epoch, + self.batch_idx + 1, + self.num_batches, + batch_time=batch_time, + data_time=data_time, + eta=eta_str, + losses=losses, + lr=self.get_current_lr() + ) + ) + + if self.writer is not None: + n_iter = self.epoch * self.num_batches + self.batch_idx + self.writer.add_scalar('Train/time', batch_time.avg, n_iter) + self.writer.add_scalar('Train/data', data_time.avg, n_iter) + for name, meter in losses.meters.items(): + self.writer.add_scalar('Train/' + name, meter.avg, n_iter) + self.writer.add_scalar( + 'Train/lr', self.get_current_lr(), n_iter + ) + + end = time.time() + + if self.batch_idx == 4: # 前5个step不记录时间 + batch_time.reset() + + self.update_lr() + return batch_time.avg + + def forward_backward(self, data): + raise NotImplementedError + + def test( + self, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + save_dir='', + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + r"""Tests model on target datasets. + + .. note:: + + This function has been called in ``run()``. + + .. note:: + + The test pipeline implemented in this function suits both image- and + video-reid. In general, a subclass of Engine only needs to re-implement + ``extract_features()`` and ``parse_data_for_eval()`` (most of the time), + but not a must. Please refer to the source code for more details. + """ + self.set_model_mode('eval') + targets = list(self.test_loader.keys()) + + for name in targets: + domain = 'source' if name in self.datamanager.sources else 'target' + print('##### Evaluating {} ({}) #####'.format(name, domain)) + query_loader = self.test_loader[name]['query'] + gallery_loader = self.test_loader[name]['gallery'] + rank1, mAP = self._evaluate( + dataset_name=name, + query_loader=query_loader, + gallery_loader=gallery_loader, + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks, + rerank=rerank + ) + + if self.writer is not None: + self.writer.add_scalar(f'Test/{name}/rank1', rank1, self.epoch) + self.writer.add_scalar(f'Test/{name}/mAP', mAP, self.epoch) + + return rank1 + + @torch.no_grad() + def _evaluate( + self, + dataset_name='', + query_loader=None, + gallery_loader=None, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + save_dir='', + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + batch_time = AverageMeter() + + def _feature_extraction(data_loader): + f_, pids_, camids_ = [], [], [] + for batch_idx, data in enumerate(data_loader): + imgs, pids, camids = self.parse_data_for_eval(data) + if self.use_gpu: + imgs = imgs.cuda() + elif self.use_npu: + imgs = imgs.npu() + end = time.time() + features = self.extract_features(imgs) + batch_time.update(time.time() - end) + features = features.cpu().clone() + f_.append(features) + pids_.extend(pids) + camids_.extend(camids) + f_ = torch.cat(f_, 0) + pids_ = np.asarray(pids_) + camids_ = np.asarray(camids_) + return f_, pids_, camids_ + + print('Extracting features from query set ...') + qf, q_pids, q_camids = _feature_extraction(query_loader) + print('Done, obtained {}-by-{} matrix'.format(qf.size(0), qf.size(1))) + + print('Extracting features from gallery set ...') + gf, g_pids, g_camids = _feature_extraction(gallery_loader) + print('Done, obtained {}-by-{} matrix'.format(gf.size(0), gf.size(1))) + + print('Speed: {:.4f} sec/batch'.format(batch_time.avg)) + + if normalize_feature: + print('Normalzing features with L2 norm ...') + qf = F.normalize(qf, p=2, dim=1) + gf = F.normalize(gf, p=2, dim=1) + + print( + 'Computing distance matrix with metric={} ...'.format(dist_metric) + ) + distmat = metrics.compute_distance_matrix(qf, gf, dist_metric) + distmat = distmat.numpy() + + if rerank: + print('Applying person re-ranking ...') + distmat_qq = metrics.compute_distance_matrix(qf, qf, dist_metric) + distmat_gg = metrics.compute_distance_matrix(gf, gf, dist_metric) + distmat = re_ranking(distmat, distmat_qq, distmat_gg) + + print('Computing CMC and mAP ...') + cmc, mAP = metrics.evaluate_rank( + distmat, + q_pids, + g_pids, + q_camids, + g_camids, + use_metric_cuhk03=use_metric_cuhk03 + ) + + print('** Results **') + print('mAP: {:.1%}'.format(mAP)) + print('CMC curve') + for r in ranks: + print('Rank-{:<3}: {:.1%}'.format(r, cmc[r - 1])) + + if visrank: + visualize_ranked_results( + distmat, + self.datamanager.fetch_test_loaders(dataset_name), + self.datamanager.data_type, + width=self.datamanager.width, + height=self.datamanager.height, + save_dir=osp.join(save_dir, 'visrank_' + dataset_name), + topk=visrank_topk + ) + + return cmc[0], mAP + + def compute_loss(self, criterion, outputs, targets): + if isinstance(outputs, (tuple, list)): + loss = DeepSupervision(criterion, outputs, targets) + else: + loss = criterion(outputs, targets) + return loss + + def extract_features(self, input): + return self.model(input) + + def parse_data_for_train(self, data): + imgs = data['img'] + pids = data['pid'] + return imgs, pids + + def parse_data_for_eval(self, data): + imgs = data['img'] + pids = data['pid'] + camids = data['camid'] + return imgs, pids, camids + + def two_stepped_transfer_learning( + self, epoch, fixbase_epoch, open_layers, model=None + ): + """Two-stepped transfer learning. + + The idea is to freeze base layers for a certain number of epochs + and then open all layers for training. + + Reference: https://arxiv.org/abs/1611.05244 + """ + model = self.model if model is None else model + if model is None: + return + + if (epoch + 1) <= fixbase_epoch and open_layers is not None: + print( + '* Only train {} (epoch: {}/{})'.format( + open_layers, epoch + 1, fixbase_epoch + ) + ) + open_specified_layers(model, open_layers) + else: + open_all_layers(model) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py new file mode 100644 index 0000000000..0376bc80de --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py @@ -0,0 +1,68 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + +from .cross_entropy_loss import CrossEntropyLoss +from .hard_mine_triplet_loss import TripletLoss + + +def DeepSupervision(criterion, xs, y): + """DeepSupervision + + Applies criterion to each element in a list. + + Args: + criterion: loss function + xs: tuple of inputs + y: ground truth + """ + loss = 0. + for x in xs: + loss += criterion(x, y) + loss /= len(xs) + return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py new file mode 100644 index 0000000000..d043691331 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py @@ -0,0 +1,100 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn + + +class CrossEntropyLoss(nn.Module): + r"""Cross entropy loss with label smoothing regularizer. + + Reference: + Szegedy et al. Rethinking the Inception Architecture for Computer Vision. CVPR 2016. + + With label smoothing, the label :math:`y` for a class is computed by + + .. math:: + \begin{equation} + (1 - \eps) \times y + \frac{\eps}{K}, + \end{equation} + + where :math:`K` denotes the number of classes and :math:`\eps` is a weight. When + :math:`\eps = 0`, the loss function reduces to the normal cross entropy. + + Args: + num_classes (int): number of classes. + eps (float, optional): weight. Default is 0.1. + use_gpu (bool, optional): whether to use gpu devices. Default is True. + label_smooth (bool, optional): whether to apply label smoothing. Default is True. + """ + + def __init__(self, num_classes, eps=0.1, use_gpu=False, use_npu=False, label_smooth=True): + super(CrossEntropyLoss, self).__init__() + self.num_classes = num_classes + self.eps = eps if label_smooth else 0 + self.use_gpu = use_gpu + self.use_npu = use_npu + self.logsoftmax = nn.LogSoftmax(dim=1) + + def forward(self, inputs, targets): + """ + Args: + inputs (torch.Tensor): prediction matrix (before softmax) with + shape (batch_size, num_classes). + targets (torch.LongTensor): ground truth labels with shape (batch_size). + Each position contains the label index. + """ + log_probs = self.logsoftmax(inputs) + zeros = torch.zeros(log_probs.size()) + targets = zeros.scatter_(1, targets.unsqueeze(1).data.cpu(), 1) + if self.use_gpu: + targets = targets.cuda() + elif self.use_npu: + targets = targets.npu() + targets = (1 - self.eps) * targets + self.eps / self.num_classes + return (-targets * log_probs).mean(0).sum() diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py new file mode 100644 index 0000000000..ac2d927c92 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py @@ -0,0 +1,95 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn + + +class TripletLoss(nn.Module): + """Triplet loss with hard positive/negative mining. + + Reference: + Hermans et al. In Defense of the Triplet Loss for Person Re-Identification. arXiv:1703.07737. + + Imported from ``_. + + Args: + margin (float, optional): margin for triplet. Default is 0.3. + """ + + def __init__(self, margin=0.3): + super(TripletLoss, self).__init__() + self.margin = margin + self.ranking_loss = nn.MarginRankingLoss(margin=margin) + + def forward(self, inputs, targets): + """ + Args: + inputs (torch.Tensor): feature matrix with shape (batch_size, feat_dim). + targets (torch.LongTensor): ground truth labels with shape (num_classes). + """ + n = inputs.size(0) + + # Compute pairwise distance, replace by the official when merged + dist = torch.pow(inputs, 2).sum(dim=1, keepdim=True).expand(n, n) + dist = dist + dist.t() + dist.addmm_(inputs, inputs.t(), beta=1, alpha=-2) + dist = dist.clamp(min=1e-12).sqrt() # for numerical stability + + # For each anchor, find the hardest positive and negative + mask = targets.expand(n, n).eq(targets.expand(n, n).t()) + dist_ap, dist_an = [], [] + for i in range(n): + dist_ap.append(dist[i][mask[i]].max().unsqueeze(0)) + dist_an.append(dist[i][mask[i] == 0].min().unsqueeze(0)) + dist_ap = torch.cat(dist_ap) + dist_an = torch.cat(dist_an) + + # Compute ranking hinge loss + y = torch.ones_like(dist_an) + return self.ranking_loss(dist_an, dist_ap, y) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py new file mode 100644 index 0000000000..b1c17830fa --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py @@ -0,0 +1,52 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .rank import evaluate_rank +from .accuracy import accuracy +from .distance import compute_distance_matrix diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py new file mode 100644 index 0000000000..c5145818b8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py @@ -0,0 +1,84 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + + +def accuracy(output, target, topk=(1, )): + """Computes the accuracy over the k top predictions for + the specified values of k. + + Args: + output (torch.Tensor): prediction matrix with shape (batch_size, num_classes). + target (torch.LongTensor): ground truth labels with shape (batch_size). + topk (tuple, optional): accuracy at top-k will be computed. For example, + topk=(1, 5) means accuracy at top-1 and top-5 will be computed. + + Returns: + list: accuracy at top-k. + + Examples:: + >>> from torchreid import metrics + >>> metrics.accuracy(output, target) + """ + maxk = max(topk) + batch_size = target.size(0) + + if isinstance(output, (tuple, list)): + output = output[0] + + _, pred = output.topk(maxk, 1, True, True) + pred = pred.t() + correct = pred.eq(target.view(1, -1).expand_as(pred)) + + res = [] + for k in topk: + correct_k = correct[:k].view(-1).float().sum(0, keepdim=True) + acc = correct_k.mul_(100.0 / batch_size) + res.append(acc) + + return res diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py new file mode 100644 index 0000000000..f209c03fa2 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py @@ -0,0 +1,127 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch +from torch.nn import functional as F + + +def compute_distance_matrix(input1, input2, metric='euclidean'): + """A wrapper function for computing distance matrix. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + metric (str, optional): "euclidean" or "cosine". + Default is "euclidean". + + Returns: + torch.Tensor: distance matrix. + + Examples:: + >>> from torchreid import metrics + >>> input1 = torch.rand(10, 2048) + >>> input2 = torch.rand(100, 2048) + >>> distmat = metrics.compute_distance_matrix(input1, input2) + >>> distmat.size() # (10, 100) + """ + # check input + assert isinstance(input1, torch.Tensor) + assert isinstance(input2, torch.Tensor) + assert input1.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( + input1.dim() + ) + assert input2.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( + input2.dim() + ) + assert input1.size(1) == input2.size(1) + + if metric == 'euclidean': + distmat = euclidean_squared_distance(input1, input2) + elif metric == 'cosine': + distmat = cosine_distance(input1, input2) + else: + raise ValueError( + 'Unknown distance metric: {}. ' + 'Please choose either "euclidean" or "cosine"'.format(metric) + ) + + return distmat + + +def euclidean_squared_distance(input1, input2): + """Computes euclidean squared distance. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + + Returns: + torch.Tensor: distance matrix. + """ + m, n = input1.size(0), input2.size(0) + mat1 = torch.pow(input1, 2).sum(dim=1, keepdim=True).expand(m, n) + mat2 = torch.pow(input2, 2).sum(dim=1, keepdim=True).expand(n, m).t() + distmat = mat1 + mat2 + distmat.addmm_(input1, input2.t(), beta=1, alpha=-2) + return distmat + + +def cosine_distance(input1, input2): + """Computes cosine distance. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + + Returns: + torch.Tensor: distance matrix. + """ + input1_normed = F.normalize(input1, p=2, dim=1) + input2_normed = F.normalize(input2, p=2, dim=1) + distmat = 1 - torch.mm(input1_normed, input2_normed.t()) + return distmat diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py new file mode 100644 index 0000000000..8e9fc70253 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py @@ -0,0 +1,254 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import numpy as np +import warnings +from collections import defaultdict + +try: + from torchreid.metrics.rank_cylib.rank_cy import evaluate_cy + IS_CYTHON_AVAI = True +except ImportError: + IS_CYTHON_AVAI = False + warnings.warn( + 'Cython evaluation (very fast so highly recommended) is ' + 'unavailable, now use python evaluation.' + ) + + +def eval_cuhk03(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): + """Evaluation with cuhk03 metric + Key: one image for each gallery identity is randomly sampled for each query identity. + Random sampling is performed num_repeats times. + """ + num_repeats = 10 + num_q, num_g = distmat.shape + + if num_g < max_rank: + max_rank = num_g + print( + 'Note: number of gallery samples is quite small, got {}'. + format(num_g) + ) + + indices = np.argsort(distmat, axis=1) + matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) + + # compute cmc curve for each query + all_cmc = [] + all_AP = [] + num_valid_q = 0. # number of valid query + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + order = indices[q_idx] + remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) + keep = np.invert(remove) + + # compute cmc curve + raw_cmc = matches[q_idx][ + keep] # binary vector, positions with value 1 are correct matches + if not np.any(raw_cmc): + # this condition is true when query identity does not appear in gallery + continue + + kept_g_pids = g_pids[order][keep] + g_pids_dict = defaultdict(list) + for idx, pid in enumerate(kept_g_pids): + g_pids_dict[pid].append(idx) + + cmc = 0. + for repeat_idx in range(num_repeats): + mask = np.zeros(len(raw_cmc), dtype=np.bool) + for _, idxs in g_pids_dict.items(): + # randomly sample one image for each gallery person + rnd_idx = np.random.choice(idxs) + mask[rnd_idx] = True + masked_raw_cmc = raw_cmc[mask] + _cmc = masked_raw_cmc.cumsum() + _cmc[_cmc > 1] = 1 + cmc += _cmc[:max_rank].astype(np.float32) + + cmc /= num_repeats + all_cmc.append(cmc) + # compute AP + num_rel = raw_cmc.sum() + tmp_cmc = raw_cmc.cumsum() + tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] + tmp_cmc = np.asarray(tmp_cmc) * raw_cmc + AP = tmp_cmc.sum() / num_rel + all_AP.append(AP) + num_valid_q += 1. + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + all_cmc = np.asarray(all_cmc).astype(np.float32) + all_cmc = all_cmc.sum(0) / num_valid_q + mAP = np.mean(all_AP) + + return all_cmc, mAP + + +def eval_market1501(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): + """Evaluation with market1501 metric + Key: for each query identity, its gallery images from the same camera view are discarded. + """ + num_q, num_g = distmat.shape + + if num_g < max_rank: + max_rank = num_g + print( + 'Note: number of gallery samples is quite small, got {}'. + format(num_g) + ) + + indices = np.argsort(distmat, axis=1) + matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) + + # compute cmc curve for each query + all_cmc = [] + all_AP = [] + num_valid_q = 0. # number of valid query + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + order = indices[q_idx] + remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) + keep = np.invert(remove) + + # compute cmc curve + raw_cmc = matches[q_idx][ + keep] # binary vector, positions with value 1 are correct matches + if not np.any(raw_cmc): + # this condition is true when query identity does not appear in gallery + continue + + cmc = raw_cmc.cumsum() + cmc[cmc > 1] = 1 + + all_cmc.append(cmc[:max_rank]) + num_valid_q += 1. + + # compute average precision + # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision + num_rel = raw_cmc.sum() + tmp_cmc = raw_cmc.cumsum() + tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] + tmp_cmc = np.asarray(tmp_cmc) * raw_cmc + AP = tmp_cmc.sum() / num_rel + all_AP.append(AP) + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + all_cmc = np.asarray(all_cmc).astype(np.float32) + all_cmc = all_cmc.sum(0) / num_valid_q + mAP = np.mean(all_AP) + + return all_cmc, mAP + + +def evaluate_py( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03 +): + if use_metric_cuhk03: + return eval_cuhk03( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank + ) + else: + return eval_market1501( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank + ) + + +def evaluate_rank( + distmat, + q_pids, + g_pids, + q_camids, + g_camids, + max_rank=50, + use_metric_cuhk03=False, + use_cython=True +): + """Evaluates CMC rank. + + Args: + distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). + q_pids (numpy.ndarray): 1-D array containing person identities + of each query instance. + g_pids (numpy.ndarray): 1-D array containing person identities + of each gallery instance. + q_camids (numpy.ndarray): 1-D array containing camera views under + which each query instance is captured. + g_camids (numpy.ndarray): 1-D array containing camera views under + which each gallery instance is captured. + max_rank (int, optional): maximum CMC rank to be computed. Default is 50. + use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. + Default is False. This should be enabled when using cuhk03 classic split. + use_cython (bool, optional): use cython code for evaluation. Default is True. + This is highly recommended as the cython code can speed up the cmc computation + by more than 10x. This requires Cython to be installed. + """ + if use_cython and IS_CYTHON_AVAI: + return evaluate_cy( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, + use_metric_cuhk03 + ) + else: + return evaluate_py( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, + use_metric_cuhk03 + ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py new file mode 100644 index 0000000000..204aea72a1 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py @@ -0,0 +1,425 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code source: https://github.com/pytorch/vision +""" +from __future__ import division, absolute_import +import re +from collections import OrderedDict +import torch +import torch.nn as nn +from torch.nn import functional as F +from torch.utils import model_zoo + +__all__ = [ + 'densenet121', 'densenet169', 'densenet201', 'densenet161', + 'densenet121_fc512' +] + +model_urls = { + 'densenet121': + 'https://download.pytorch.org/models/densenet121-a639ec97.pth', + 'densenet169': + 'https://download.pytorch.org/models/densenet169-b2777c0a.pth', + 'densenet201': + 'https://download.pytorch.org/models/densenet201-c1103571.pth', + 'densenet161': + 'https://download.pytorch.org/models/densenet161-8d451a50.pth', +} + + +class _DenseLayer(nn.Sequential): + + def __init__(self, num_input_features, growth_rate, bn_size, drop_rate): + super(_DenseLayer, self).__init__() + self.add_module('norm1', nn.BatchNorm2d(num_input_features)), + self.add_module('relu1', nn.ReLU(inplace=True)), + self.add_module( + 'conv1', + nn.Conv2d( + num_input_features, + bn_size * growth_rate, + kernel_size=1, + stride=1, + bias=False + ) + ), + self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate)), + self.add_module('relu2', nn.ReLU(inplace=True)), + self.add_module( + 'conv2', + nn.Conv2d( + bn_size * growth_rate, + growth_rate, + kernel_size=3, + stride=1, + padding=1, + bias=False + ) + ), + self.drop_rate = drop_rate + + def forward(self, x): + new_features = super(_DenseLayer, self).forward(x) + if self.drop_rate > 0: + new_features = F.dropout( + new_features, p=self.drop_rate, training=self.training + ) + return torch.cat([x, new_features], 1) + + +class _DenseBlock(nn.Sequential): + + def __init__( + self, num_layers, num_input_features, bn_size, growth_rate, drop_rate + ): + super(_DenseBlock, self).__init__() + for i in range(num_layers): + layer = _DenseLayer( + num_input_features + i*growth_rate, growth_rate, bn_size, + drop_rate + ) + self.add_module('denselayer%d' % (i+1), layer) + + +class _Transition(nn.Sequential): + + def __init__(self, num_input_features, num_output_features): + super(_Transition, self).__init__() + self.add_module('norm', nn.BatchNorm2d(num_input_features)) + self.add_module('relu', nn.ReLU(inplace=True)) + self.add_module( + 'conv', + nn.Conv2d( + num_input_features, + num_output_features, + kernel_size=1, + stride=1, + bias=False + ) + ) + self.add_module('pool', nn.AvgPool2d(kernel_size=2, stride=2)) + + +class DenseNet(nn.Module): + """Densely connected network. + + Reference: + Huang et al. Densely Connected Convolutional Networks. CVPR 2017. + + Public keys: + - ``densenet121``: DenseNet121. + - ``densenet169``: DenseNet169. + - ``densenet201``: DenseNet201. + - ``densenet161``: DenseNet161. + - ``densenet121_fc512``: DenseNet121 + FC. + """ + + def __init__( + self, + num_classes, + loss, + growth_rate=32, + block_config=(6, 12, 24, 16), + num_init_features=64, + bn_size=4, + drop_rate=0, + fc_dims=None, + dropout_p=None, + **kwargs + ): + + super(DenseNet, self).__init__() + self.loss = loss + + # First convolution + self.features = nn.Sequential( + OrderedDict( + [ + ( + 'conv0', + nn.Conv2d( + 3, + num_init_features, + kernel_size=7, + stride=2, + padding=3, + bias=False + ) + ), + ('norm0', nn.BatchNorm2d(num_init_features)), + ('relu0', nn.ReLU(inplace=True)), + ( + 'pool0', + nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + ), + ] + ) + ) + + # Each denseblock + num_features = num_init_features + for i, num_layers in enumerate(block_config): + block = _DenseBlock( + num_layers=num_layers, + num_input_features=num_features, + bn_size=bn_size, + growth_rate=growth_rate, + drop_rate=drop_rate + ) + self.features.add_module('denseblock%d' % (i+1), block) + num_features = num_features + num_layers*growth_rate + if i != len(block_config) - 1: + trans = _Transition( + num_input_features=num_features, + num_output_features=num_features // 2 + ) + self.features.add_module('transition%d' % (i+1), trans) + num_features = num_features // 2 + + # Final batch norm + self.features.add_module('norm5', nn.BatchNorm2d(num_features)) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.feature_dim = num_features + self.fc = self._construct_fc_layer(fc_dims, num_features, dropout_p) + + # Linear layer + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer. + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def forward(self, x): + f = self.features(x) + f = F.relu(f, inplace=True) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + + # '.'s are no longer allowed in module names, but pervious _DenseLayer + # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. + # They are also in the checkpoints in model_urls. This pattern is used + # to find such keys. + pattern = re.compile( + r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$' + ) + for key in list(pretrain_dict.keys()): + res = pattern.match(key) + if res: + new_key = res.group(1) + res.group(2) + pretrain_dict[new_key] = pretrain_dict[key] + del pretrain_dict[key] + + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +""" +Dense network configurations: +-- +densenet121: num_init_features=64, growth_rate=32, block_config=(6, 12, 24, 16) +densenet169: num_init_features=64, growth_rate=32, block_config=(6, 12, 32, 32) +densenet201: num_init_features=64, growth_rate=32, block_config=(6, 12, 48, 32) +densenet161: num_init_features=96, growth_rate=48, block_config=(6, 12, 36, 24) +""" + + +def densenet121(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 24, 16), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet121']) + return model + + +def densenet169(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 32, 32), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet169']) + return model + + +def densenet201(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 48, 32), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet201']) + return model + + +def densenet161(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=96, + growth_rate=48, + block_config=(6, 12, 36, 24), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet161']) + return model + + +def densenet121_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 24, 16), + fc_dims=[512], + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet121']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py new file mode 100644 index 0000000000..27dae2fa28 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py @@ -0,0 +1,461 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +from torch import nn +from torch.nn import functional as F + +__all__ = ['HACNN'] + + +class ConvBlock(nn.Module): + """Basic convolutional block. + + convolution + batch normalization + relu. + + Args: + in_c (int): number of input channels. + out_c (int): number of output channels. + k (int or tuple): kernel size. + s (int or tuple): stride. + p (int or tuple): padding. + """ + + def __init__(self, in_c, out_c, k, s=1, p=0): + super(ConvBlock, self).__init__() + self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) + self.bn = nn.BatchNorm2d(out_c) + + def forward(self, x): + return F.relu(self.bn(self.conv(x))) + + +class InceptionA(nn.Module): + + def __init__(self, in_channels, out_channels): + super(InceptionA, self).__init__() + mid_channels = out_channels // 4 + + self.stream1 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream2 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream3 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream4 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1), + ConvBlock(in_channels, mid_channels, 1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + s4 = self.stream4(x) + y = torch.cat([s1, s2, s3, s4], dim=1) + return y + + +class InceptionB(nn.Module): + + def __init__(self, in_channels, out_channels): + super(InceptionB, self).__init__() + mid_channels = out_channels // 4 + + self.stream1 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), + ) + self.stream2 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), + ) + self.stream3 = nn.Sequential( + nn.MaxPool2d(3, stride=2, padding=1), + ConvBlock(in_channels, mid_channels * 2, 1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + y = torch.cat([s1, s2, s3], dim=1) + return y + + +class SpatialAttn(nn.Module): + """Spatial Attention (Sec. 3.1.I.1)""" + + def __init__(self): + super(SpatialAttn, self).__init__() + self.conv1 = ConvBlock(1, 1, 3, s=2, p=1) + self.conv2 = ConvBlock(1, 1, 1) + + def forward(self, x): + # global cross-channel averaging + x = x.mean(1, keepdim=True) + # 3-by-3 conv + x = self.conv1(x) + # bilinear resizing + x = F.upsample( + x, (x.size(2) * 2, x.size(3) * 2), + mode='bilinear', + align_corners=True + ) + # scaling conv + x = self.conv2(x) + return x + + +class ChannelAttn(nn.Module): + """Channel Attention (Sec. 3.1.I.2)""" + + def __init__(self, in_channels, reduction_rate=16): + super(ChannelAttn, self).__init__() + assert in_channels % reduction_rate == 0 + self.conv1 = ConvBlock(in_channels, in_channels // reduction_rate, 1) + self.conv2 = ConvBlock(in_channels // reduction_rate, in_channels, 1) + + def forward(self, x): + # squeeze operation (global average pooling) + x = F.avg_pool2d(x, x.size()[2:]) + # excitation operation (2 conv layers) + x = self.conv1(x) + x = self.conv2(x) + return x + + +class SoftAttn(nn.Module): + """Soft Attention (Sec. 3.1.I) + + Aim: Spatial Attention + Channel Attention + + Output: attention maps with shape identical to input. + """ + + def __init__(self, in_channels): + super(SoftAttn, self).__init__() + self.spatial_attn = SpatialAttn() + self.channel_attn = ChannelAttn(in_channels) + self.conv = ConvBlock(in_channels, in_channels, 1) + + def forward(self, x): + y_spatial = self.spatial_attn(x) + y_channel = self.channel_attn(x) + y = y_spatial * y_channel + y = torch.sigmoid(self.conv(y)) + return y + + +class HardAttn(nn.Module): + """Hard Attention (Sec. 3.1.II)""" + + def __init__(self, in_channels): + super(HardAttn, self).__init__() + self.fc = nn.Linear(in_channels, 4 * 2) + self.init_params() + + def init_params(self): + self.fc.weight.data.zero_() + self.fc.bias.data.copy_( + torch.tensor( + [0, -0.75, 0, -0.25, 0, 0.25, 0, 0.75], dtype=torch.float + ) + ) + + def forward(self, x): + # squeeze operation (global average pooling) + x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), x.size(1)) + # predict transformation parameters + theta = torch.tanh(self.fc(x)) + theta = theta.view(-1, 4, 2) + return theta + + +class HarmAttn(nn.Module): + """Harmonious Attention (Sec. 3.1)""" + + def __init__(self, in_channels): + super(HarmAttn, self).__init__() + self.soft_attn = SoftAttn(in_channels) + self.hard_attn = HardAttn(in_channels) + + def forward(self, x): + y_soft_attn = self.soft_attn(x) + theta = self.hard_attn(x) + return y_soft_attn, theta + + +class HACNN(nn.Module): + """Harmonious Attention Convolutional Neural Network. + + Reference: + Li et al. Harmonious Attention Network for Person Re-identification. CVPR 2018. + + Public keys: + - ``hacnn``: HACNN. + """ + + # Args: + # num_classes (int): number of classes to predict + # nchannels (list): number of channels AFTER concatenation + # feat_dim (int): feature dimension for a single stream + # learn_region (bool): whether to learn region features (i.e. local branch) + + def __init__( + self, + num_classes, + loss='softmax', + nchannels=[128, 256, 384], + feat_dim=512, + learn_region=True, + use_gpu=True, + **kwargs + ): + super(HACNN, self).__init__() + self.loss = loss + self.learn_region = learn_region + self.use_gpu = use_gpu + + self.conv = ConvBlock(3, 32, 3, s=2, p=1) + + # Construct Inception + HarmAttn blocks + # ============== Block 1 ============== + self.inception1 = nn.Sequential( + InceptionA(32, nchannels[0]), + InceptionB(nchannels[0], nchannels[0]), + ) + self.ha1 = HarmAttn(nchannels[0]) + + # ============== Block 2 ============== + self.inception2 = nn.Sequential( + InceptionA(nchannels[0], nchannels[1]), + InceptionB(nchannels[1], nchannels[1]), + ) + self.ha2 = HarmAttn(nchannels[1]) + + # ============== Block 3 ============== + self.inception3 = nn.Sequential( + InceptionA(nchannels[1], nchannels[2]), + InceptionB(nchannels[2], nchannels[2]), + ) + self.ha3 = HarmAttn(nchannels[2]) + + self.fc_global = nn.Sequential( + nn.Linear(nchannels[2], feat_dim), + nn.BatchNorm1d(feat_dim), + nn.ReLU(), + ) + self.classifier_global = nn.Linear(feat_dim, num_classes) + + if self.learn_region: + self.init_scale_factors() + self.local_conv1 = InceptionB(32, nchannels[0]) + self.local_conv2 = InceptionB(nchannels[0], nchannels[1]) + self.local_conv3 = InceptionB(nchannels[1], nchannels[2]) + self.fc_local = nn.Sequential( + nn.Linear(nchannels[2] * 4, feat_dim), + nn.BatchNorm1d(feat_dim), + nn.ReLU(), + ) + self.classifier_local = nn.Linear(feat_dim, num_classes) + self.feat_dim = feat_dim * 2 + else: + self.feat_dim = feat_dim + + def init_scale_factors(self): + # initialize scale factors (s_w, s_h) for four regions + self.scale_factors = [] + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + + def stn(self, x, theta): + """Performs spatial transform + + x: (batch, channel, height, width) + theta: (batch, 2, 3) + """ + grid = F.affine_grid(theta, x.size()) + x = F.grid_sample(x, grid) + return x + + def transform_theta(self, theta_i, region_idx): + """Transforms theta to include (s_w, s_h), resulting in (batch, 2, 3)""" + scale_factors = self.scale_factors[region_idx] + theta = torch.zeros(theta_i.size(0), 2, 3) + theta[:, :, :2] = scale_factors + theta[:, :, -1] = theta_i + if self.use_gpu: + theta = theta.cuda() + return theta + + def forward(self, x): + assert x.size(2) == 160 and x.size(3) == 64, \ + 'Input size does not match, expected (160, 64) but got ({}, {})'.format(x.size(2), x.size(3)) + x = self.conv(x) + + # ============== Block 1 ============== + # global branch + x1 = self.inception1(x) + x1_attn, x1_theta = self.ha1(x1) + x1_out = x1 * x1_attn + # local branch + if self.learn_region: + x1_local_list = [] + for region_idx in range(4): + x1_theta_i = x1_theta[:, region_idx, :] + x1_theta_i = self.transform_theta(x1_theta_i, region_idx) + x1_trans_i = self.stn(x, x1_theta_i) + x1_trans_i = F.upsample( + x1_trans_i, (24, 28), mode='bilinear', align_corners=True + ) + x1_local_i = self.local_conv1(x1_trans_i) + x1_local_list.append(x1_local_i) + + # ============== Block 2 ============== + # Block 2 + # global branch + x2 = self.inception2(x1_out) + x2_attn, x2_theta = self.ha2(x2) + x2_out = x2 * x2_attn + # local branch + if self.learn_region: + x2_local_list = [] + for region_idx in range(4): + x2_theta_i = x2_theta[:, region_idx, :] + x2_theta_i = self.transform_theta(x2_theta_i, region_idx) + x2_trans_i = self.stn(x1_out, x2_theta_i) + x2_trans_i = F.upsample( + x2_trans_i, (12, 14), mode='bilinear', align_corners=True + ) + x2_local_i = x2_trans_i + x1_local_list[region_idx] + x2_local_i = self.local_conv2(x2_local_i) + x2_local_list.append(x2_local_i) + + # ============== Block 3 ============== + # Block 3 + # global branch + x3 = self.inception3(x2_out) + x3_attn, x3_theta = self.ha3(x3) + x3_out = x3 * x3_attn + # local branch + if self.learn_region: + x3_local_list = [] + for region_idx in range(4): + x3_theta_i = x3_theta[:, region_idx, :] + x3_theta_i = self.transform_theta(x3_theta_i, region_idx) + x3_trans_i = self.stn(x2_out, x3_theta_i) + x3_trans_i = F.upsample( + x3_trans_i, (6, 7), mode='bilinear', align_corners=True + ) + x3_local_i = x3_trans_i + x2_local_list[region_idx] + x3_local_i = self.local_conv3(x3_local_i) + x3_local_list.append(x3_local_i) + + # ============== Feature generation ============== + # global branch + x_global = F.avg_pool2d(x3_out, + x3_out.size()[2:] + ).view(x3_out.size(0), x3_out.size(1)) + x_global = self.fc_global(x_global) + # local branch + if self.learn_region: + x_local_list = [] + for region_idx in range(4): + x_local_i = x3_local_list[region_idx] + x_local_i = F.avg_pool2d(x_local_i, + x_local_i.size()[2:] + ).view(x_local_i.size(0), -1) + x_local_list.append(x_local_i) + x_local = torch.cat(x_local_list, 1) + x_local = self.fc_local(x_local) + + if not self.training: + # l2 normalization before concatenation + if self.learn_region: + x_global = x_global / x_global.norm(p=2, dim=1, keepdim=True) + x_local = x_local / x_local.norm(p=2, dim=1, keepdim=True) + return torch.cat([x_global, x_local], 1) + else: + return x_global + + prelogits_global = self.classifier_global(x_global) + if self.learn_region: + prelogits_local = self.classifier_local(x_local) + + if self.loss == 'softmax': + if self.learn_region: + return (prelogits_global, prelogits_local) + else: + return prelogits_global + + elif self.loss == 'triplet': + if self.learn_region: + return (prelogits_global, prelogits_local), (x_global, x_local) + else: + return prelogits_global, x_global + + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py new file mode 100644 index 0000000000..f9d62718b3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py @@ -0,0 +1,406 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['inceptionresnetv2'] + +pretrained_settings = { + 'inceptionresnetv2': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000 + }, + 'imagenet+background': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1001 + } + } +} + + +class BasicConv2d(nn.Module): + + def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): + super(BasicConv2d, self).__init__() + self.conv = nn.Conv2d( + in_planes, + out_planes, + kernel_size=kernel_size, + stride=stride, + padding=padding, + bias=False + ) # verify bias false + self.bn = nn.BatchNorm2d( + out_planes, + eps=0.001, # value found in tensorflow + momentum=0.1, # default pytorch value + affine=True + ) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Mixed_5b(nn.Module): + + def __init__(self): + super(Mixed_5b, self).__init__() + + self.branch0 = BasicConv2d(192, 96, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(192, 48, kernel_size=1, stride=1), + BasicConv2d(48, 64, kernel_size=5, stride=1, padding=2) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(192, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), + BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(192, 64, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Block35(nn.Module): + + def __init__(self, scale=1.0): + super(Block35, self).__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(320, 32, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(320, 32, kernel_size=1, stride=1), + BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(320, 32, kernel_size=1, stride=1), + BasicConv2d(32, 48, kernel_size=3, stride=1, padding=1), + BasicConv2d(48, 64, kernel_size=3, stride=1, padding=1) + ) + + self.conv2d = nn.Conv2d(128, 320, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Mixed_6a(nn.Module): + + def __init__(self): + super(Mixed_6a, self).__init__() + + self.branch0 = BasicConv2d(320, 384, kernel_size=3, stride=2) + + self.branch1 = nn.Sequential( + BasicConv2d(320, 256, kernel_size=1, stride=1), + BasicConv2d(256, 256, kernel_size=3, stride=1, padding=1), + BasicConv2d(256, 384, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Block17(nn.Module): + + def __init__(self, scale=1.0): + super(Block17, self).__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(1088, 192, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(1088, 128, kernel_size=1, stride=1), + BasicConv2d( + 128, 160, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 160, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) + ) + ) + + self.conv2d = nn.Conv2d(384, 1088, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Mixed_7a(nn.Module): + + def __init__(self): + super(Mixed_7a, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 384, kernel_size=3, stride=2) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 288, kernel_size=3, stride=2) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 288, kernel_size=3, stride=1, padding=1), + BasicConv2d(288, 320, kernel_size=3, stride=2) + ) + + self.branch3 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Block8(nn.Module): + + def __init__(self, scale=1.0, noReLU=False): + super(Block8, self).__init__() + + self.scale = scale + self.noReLU = noReLU + + self.branch0 = BasicConv2d(2080, 192, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(2080, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 224, kernel_size=(1, 3), stride=1, padding=(0, 1) + ), + BasicConv2d( + 224, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + ) + + self.conv2d = nn.Conv2d(448, 2080, kernel_size=1, stride=1) + if not self.noReLU: + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + if not self.noReLU: + out = self.relu(out) + return out + + +# ---------------- +# Model Definition +# ---------------- +class InceptionResNetV2(nn.Module): + """Inception-ResNet-V2. + + Reference: + Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual + Connections on Learning. AAAI 2017. + + Public keys: + - ``inceptionresnetv2``: Inception-ResNet-V2. + """ + + def __init__(self, num_classes, loss='softmax', **kwargs): + super(InceptionResNetV2, self).__init__() + self.loss = loss + + # Modules + self.conv2d_1a = BasicConv2d(3, 32, kernel_size=3, stride=2) + self.conv2d_2a = BasicConv2d(32, 32, kernel_size=3, stride=1) + self.conv2d_2b = BasicConv2d( + 32, 64, kernel_size=3, stride=1, padding=1 + ) + self.maxpool_3a = nn.MaxPool2d(3, stride=2) + self.conv2d_3b = BasicConv2d(64, 80, kernel_size=1, stride=1) + self.conv2d_4a = BasicConv2d(80, 192, kernel_size=3, stride=1) + self.maxpool_5a = nn.MaxPool2d(3, stride=2) + self.mixed_5b = Mixed_5b() + self.repeat = nn.Sequential( + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17) + ) + self.mixed_6a = Mixed_6a() + self.repeat_1 = nn.Sequential( + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10) + ) + self.mixed_7a = Mixed_7a() + self.repeat_2 = nn.Sequential( + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20) + ) + + self.block8 = Block8(noReLU=True) + self.conv2d_7b = BasicConv2d(2080, 1536, kernel_size=1, stride=1) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.classifier = nn.Linear(1536, num_classes) + + def load_imagenet_weights(self): + settings = pretrained_settings['inceptionresnetv2']['imagenet'] + pretrain_dict = model_zoo.load_url(settings['url']) + model_dict = self.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + self.load_state_dict(model_dict) + + def featuremaps(self, x): + x = self.conv2d_1a(x) + x = self.conv2d_2a(x) + x = self.conv2d_2b(x) + x = self.maxpool_3a(x) + x = self.conv2d_3b(x) + x = self.conv2d_4a(x) + x = self.maxpool_5a(x) + x = self.mixed_5b(x) + x = self.repeat(x) + x = self.mixed_6a(x) + x = self.repeat_1(x) + x = self.mixed_7a(x) + x = self.repeat_2(x) + x = self.block8(x) + x = self.conv2d_7b(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def inceptionresnetv2(num_classes, loss='softmax', pretrained=True, **kwargs): + model = InceptionResNetV2(num_classes=num_classes, loss=loss, **kwargs) + if pretrained: + model.load_imagenet_weights() + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py new file mode 100644 index 0000000000..32a847a88f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py @@ -0,0 +1,428 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['inceptionv4'] +""" +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" + +pretrained_settings = { + 'inceptionv4': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000 + }, + 'imagenet+background': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1001 + } + } +} + + +class BasicConv2d(nn.Module): + + def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): + super(BasicConv2d, self).__init__() + self.conv = nn.Conv2d( + in_planes, + out_planes, + kernel_size=kernel_size, + stride=stride, + padding=padding, + bias=False + ) # verify bias false + self.bn = nn.BatchNorm2d( + out_planes, + eps=0.001, # value found in tensorflow + momentum=0.1, # default pytorch value + affine=True + ) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Mixed_3a(nn.Module): + + def __init__(self): + super(Mixed_3a, self).__init__() + self.maxpool = nn.MaxPool2d(3, stride=2) + self.conv = BasicConv2d(64, 96, kernel_size=3, stride=2) + + def forward(self, x): + x0 = self.maxpool(x) + x1 = self.conv(x) + out = torch.cat((x0, x1), 1) + return out + + +class Mixed_4a(nn.Module): + + def __init__(self): + super(Mixed_4a, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(160, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(160, 64, kernel_size=1, stride=1), + BasicConv2d(64, 64, kernel_size=(1, 7), stride=1, padding=(0, 3)), + BasicConv2d(64, 64, kernel_size=(7, 1), stride=1, padding=(3, 0)), + BasicConv2d(64, 96, kernel_size=(3, 3), stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + return out + + +class Mixed_5a(nn.Module): + + def __init__(self): + super(Mixed_5a, self).__init__() + self.conv = BasicConv2d(192, 192, kernel_size=3, stride=2) + self.maxpool = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.conv(x) + x1 = self.maxpool(x) + out = torch.cat((x0, x1), 1) + return out + + +class Inception_A(nn.Module): + + def __init__(self): + super(Inception_A, self).__init__() + self.branch0 = BasicConv2d(384, 96, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(384, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(384, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), + BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(384, 96, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Reduction_A(nn.Module): + + def __init__(self): + super(Reduction_A, self).__init__() + self.branch0 = BasicConv2d(384, 384, kernel_size=3, stride=2) + + self.branch1 = nn.Sequential( + BasicConv2d(384, 192, kernel_size=1, stride=1), + BasicConv2d(192, 224, kernel_size=3, stride=1, padding=1), + BasicConv2d(224, 256, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Inception_B(nn.Module): + + def __init__(self): + super(Inception_B, self).__init__() + self.branch0 = BasicConv2d(1024, 384, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 224, 256, kernel_size=(7, 1), stride=1, padding=(3, 0) + ) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), + BasicConv2d( + 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 224, 224, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), + BasicConv2d( + 224, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) + ) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(1024, 128, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Reduction_B(nn.Module): + + def __init__(self): + super(Reduction_B, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d(192, 192, kernel_size=3, stride=2) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(1024, 256, kernel_size=1, stride=1), + BasicConv2d( + 256, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 256, 320, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), BasicConv2d(320, 320, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Inception_C(nn.Module): + + def __init__(self): + super(Inception_C, self).__init__() + + self.branch0 = BasicConv2d(1536, 256, kernel_size=1, stride=1) + + self.branch1_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) + self.branch1_1a = BasicConv2d( + 384, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch1_1b = BasicConv2d( + 384, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + + self.branch2_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) + self.branch2_1 = BasicConv2d( + 384, 448, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + self.branch2_2 = BasicConv2d( + 448, 512, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch2_3a = BasicConv2d( + 512, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch2_3b = BasicConv2d( + 512, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(1536, 256, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + + x1_0 = self.branch1_0(x) + x1_1a = self.branch1_1a(x1_0) + x1_1b = self.branch1_1b(x1_0) + x1 = torch.cat((x1_1a, x1_1b), 1) + + x2_0 = self.branch2_0(x) + x2_1 = self.branch2_1(x2_0) + x2_2 = self.branch2_2(x2_1) + x2_3a = self.branch2_3a(x2_2) + x2_3b = self.branch2_3b(x2_2) + x2 = torch.cat((x2_3a, x2_3b), 1) + + x3 = self.branch3(x) + + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class InceptionV4(nn.Module): + """Inception-v4. + + Reference: + Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual + Connections on Learning. AAAI 2017. + + Public keys: + - ``inceptionv4``: InceptionV4. + """ + + def __init__(self, num_classes, loss, **kwargs): + super(InceptionV4, self).__init__() + self.loss = loss + + self.features = nn.Sequential( + BasicConv2d(3, 32, kernel_size=3, stride=2), + BasicConv2d(32, 32, kernel_size=3, stride=1), + BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1), + Mixed_3a(), + Mixed_4a(), + Mixed_5a(), + Inception_A(), + Inception_A(), + Inception_A(), + Inception_A(), + Reduction_A(), # Mixed_6a + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Reduction_B(), # Mixed_7a + Inception_C(), + Inception_C(), + Inception_C() + ) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.classifier = nn.Linear(1536, num_classes) + + def forward(self, x): + f = self.features(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def inceptionv4(num_classes, loss='softmax', pretrained=True, **kwargs): + model = InceptionV4(num_classes, loss, **kwargs) + if pretrained: + model_url = pretrained_settings['inceptionv4']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py new file mode 100644 index 0000000000..0e538241f7 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py @@ -0,0 +1,316 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['mlfn'] + +model_urls = { + # training epoch = 5, top1 = 51.6 + 'imagenet': + 'https://mega.nz/#!YHxAhaxC!yu9E6zWl0x5zscSouTdbZu8gdFFytDdl-RAdD2DEfpk', +} + + +class MLFNBlock(nn.Module): + + def __init__( + self, in_channels, out_channels, stride, fsm_channels, groups=32 + ): + super(MLFNBlock, self).__init__() + self.groups = groups + mid_channels = out_channels // 2 + + # Factor Modules + self.fm_conv1 = nn.Conv2d(in_channels, mid_channels, 1, bias=False) + self.fm_bn1 = nn.BatchNorm2d(mid_channels) + self.fm_conv2 = nn.Conv2d( + mid_channels, + mid_channels, + 3, + stride=stride, + padding=1, + bias=False, + groups=self.groups + ) + self.fm_bn2 = nn.BatchNorm2d(mid_channels) + self.fm_conv3 = nn.Conv2d(mid_channels, out_channels, 1, bias=False) + self.fm_bn3 = nn.BatchNorm2d(out_channels) + + # Factor Selection Module + self.fsm = nn.Sequential( + nn.AdaptiveAvgPool2d(1), + nn.Conv2d(in_channels, fsm_channels[0], 1), + nn.BatchNorm2d(fsm_channels[0]), + nn.ReLU(inplace=True), + nn.Conv2d(fsm_channels[0], fsm_channels[1], 1), + nn.BatchNorm2d(fsm_channels[1]), + nn.ReLU(inplace=True), + nn.Conv2d(fsm_channels[1], self.groups, 1), + nn.BatchNorm2d(self.groups), + nn.Sigmoid(), + ) + + self.downsample = None + if in_channels != out_channels or stride > 1: + self.downsample = nn.Sequential( + nn.Conv2d( + in_channels, out_channels, 1, stride=stride, bias=False + ), + nn.BatchNorm2d(out_channels), + ) + + def forward(self, x): + residual = x + s = self.fsm(x) + + # reduce dimension + x = self.fm_conv1(x) + x = self.fm_bn1(x) + x = F.relu(x, inplace=True) + + # group convolution + x = self.fm_conv2(x) + x = self.fm_bn2(x) + x = F.relu(x, inplace=True) + + # factor selection + b, c = x.size(0), x.size(1) + n = c // self.groups + ss = s.repeat(1, n, 1, 1) # from (b, g, 1, 1) to (b, g*n=c, 1, 1) + ss = ss.view(b, n, self.groups, 1, 1) + ss = ss.permute(0, 2, 1, 3, 4).contiguous() + ss = ss.view(b, c, 1, 1) + x = ss * x + + # recover dimension + x = self.fm_conv3(x) + x = self.fm_bn3(x) + x = F.relu(x, inplace=True) + + if self.downsample is not None: + residual = self.downsample(residual) + + return F.relu(residual + x, inplace=True), s + + +class MLFN(nn.Module): + """Multi-Level Factorisation Net. + + Reference: + Chang et al. Multi-Level Factorisation Net for + Person Re-Identification. CVPR 2018. + + Public keys: + - ``mlfn``: MLFN (Multi-Level Factorisation Net). + """ + + def __init__( + self, + num_classes, + loss='softmax', + groups=32, + channels=[64, 256, 512, 1024, 2048], + embed_dim=1024, + **kwargs + ): + super(MLFN, self).__init__() + self.loss = loss + self.groups = groups + + # first convolutional layer + self.conv1 = nn.Conv2d(3, channels[0], 7, stride=2, padding=3) + self.bn1 = nn.BatchNorm2d(channels[0]) + self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) + + # main body + self.feature = nn.ModuleList( + [ + # layer 1-3 + MLFNBlock(channels[0], channels[1], 1, [128, 64], self.groups), + MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), + MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), + # layer 4-7 + MLFNBlock( + channels[1], channels[2], 2, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + # layer 8-13 + MLFNBlock( + channels[2], channels[3], 2, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + # layer 14-16 + MLFNBlock( + channels[3], channels[4], 2, [512, 128], self.groups + ), + MLFNBlock( + channels[4], channels[4], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[4], channels[4], 1, [512, 128], self.groups + ), + ] + ) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + + # projection functions + self.fc_x = nn.Sequential( + nn.Conv2d(channels[4], embed_dim, 1, bias=False), + nn.BatchNorm2d(embed_dim), + nn.ReLU(inplace=True), + ) + self.fc_s = nn.Sequential( + nn.Conv2d(self.groups * 16, embed_dim, 1, bias=False), + nn.BatchNorm2d(embed_dim), + nn.ReLU(inplace=True), + ) + + self.classifier = nn.Linear(embed_dim, num_classes) + + self.init_params() + + def init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def forward(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = F.relu(x, inplace=True) + x = self.maxpool(x) + + s_hat = [] + for block in self.feature: + x, s = block(x) + s_hat.append(s) + s_hat = torch.cat(s_hat, 1) + + x = self.global_avgpool(x) + x = self.fc_x(x) + s_hat = self.fc_s(s_hat) + + v = (x+s_hat) * 0.5 + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def mlfn(num_classes, loss='softmax', pretrained=True, **kwargs): + model = MLFN(num_classes, loss, **kwargs) + if pretrained: + # init_pretrained_weights(model, model_urls['imagenet']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['imagenet']) + ) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py new file mode 100644 index 0000000000..690dade1bb --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py @@ -0,0 +1,321 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['mobilenetv2_x1_0', 'mobilenetv2_x1_4'] + +model_urls = { + # 1.0: top-1 71.3 + 'mobilenetv2_x1_0': + 'https://mega.nz/#!NKp2wAIA!1NH1pbNzY_M2hVk_hdsxNM1NUOWvvGPHhaNr-fASF6c', + # 1.4: top-1 73.9 + 'mobilenetv2_x1_4': + 'https://mega.nz/#!RGhgEIwS!xN2s2ZdyqI6vQ3EwgmRXLEW3khr9tpXg96G9SUJugGk', +} + + +class ConvBlock(nn.Module): + """Basic convolutional block. + + convolution (bias discarded) + batch normalization + relu6. + + Args: + in_c (int): number of input channels. + out_c (int): number of output channels. + k (int or tuple): kernel size. + s (int or tuple): stride. + p (int or tuple): padding. + g (int): number of blocked connections from input channels + to output channels (default: 1). + """ + + def __init__(self, in_c, out_c, k, s=1, p=0, g=1): + super(ConvBlock, self).__init__() + self.conv = nn.Conv2d( + in_c, out_c, k, stride=s, padding=p, bias=False, groups=g + ) + self.bn = nn.BatchNorm2d(out_c) + + def forward(self, x): + return F.relu6(self.bn(self.conv(x))) + + +class Bottleneck(nn.Module): + + def __init__(self, in_channels, out_channels, expansion_factor, stride=1): + super(Bottleneck, self).__init__() + mid_channels = in_channels * expansion_factor + self.use_residual = stride == 1 and in_channels == out_channels + self.conv1 = ConvBlock(in_channels, mid_channels, 1) + self.dwconv2 = ConvBlock( + mid_channels, mid_channels, 3, stride, 1, g=mid_channels + ) + self.conv3 = nn.Sequential( + nn.Conv2d(mid_channels, out_channels, 1, bias=False), + nn.BatchNorm2d(out_channels), + ) + + def forward(self, x): + m = self.conv1(x) + m = self.dwconv2(m) + m = self.conv3(m) + if self.use_residual: + return x + m + else: + return m + + +class MobileNetV2(nn.Module): + """MobileNetV2. + + Reference: + Sandler et al. MobileNetV2: Inverted Residuals and + Linear Bottlenecks. CVPR 2018. + + Public keys: + - ``mobilenetv2_x1_0``: MobileNetV2 x1.0. + - ``mobilenetv2_x1_4``: MobileNetV2 x1.4. + """ + + def __init__( + self, + num_classes, + width_mult=1, + loss='softmax', + fc_dims=None, + dropout_p=None, + **kwargs + ): + super(MobileNetV2, self).__init__() + self.loss = loss + self.in_channels = int(32 * width_mult) + self.feature_dim = int(1280 * width_mult) if width_mult > 1 else 1280 + + # construct layers + self.conv1 = ConvBlock(3, self.in_channels, 3, s=2, p=1) + self.conv2 = self._make_layer( + Bottleneck, 1, int(16 * width_mult), 1, 1 + ) + self.conv3 = self._make_layer( + Bottleneck, 6, int(24 * width_mult), 2, 2 + ) + self.conv4 = self._make_layer( + Bottleneck, 6, int(32 * width_mult), 3, 2 + ) + self.conv5 = self._make_layer( + Bottleneck, 6, int(64 * width_mult), 4, 2 + ) + self.conv6 = self._make_layer( + Bottleneck, 6, int(96 * width_mult), 3, 1 + ) + self.conv7 = self._make_layer( + Bottleneck, 6, int(160 * width_mult), 3, 2 + ) + self.conv8 = self._make_layer( + Bottleneck, 6, int(320 * width_mult), 1, 1 + ) + self.conv9 = ConvBlock(self.in_channels, self.feature_dim, 1) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc = self._construct_fc_layer( + fc_dims, self.feature_dim, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _make_layer(self, block, t, c, n, s): + # t: expansion factor + # c: output channels + # n: number of blocks + # s: stride for first layer + layers = [] + layers.append(block(self.in_channels, c, t, s)) + self.in_channels = c + for i in range(1, n): + layers.append(block(self.in_channels, c, t)) + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer. + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.conv2(x) + x = self.conv3(x) + x = self.conv4(x) + x = self.conv5(x) + x = self.conv6(x) + x = self.conv7(x) + x = self.conv8(x) + x = self.conv9(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def mobilenetv2_x1_0(num_classes, loss, pretrained=True, **kwargs): + model = MobileNetV2( + num_classes, + loss=loss, + width_mult=1, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + # init_pretrained_weights(model, model_urls['mobilenetv2_x1_0']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['mobilenetv2_x1_0']) + ) + return model + + +def mobilenetv2_x1_4(num_classes, loss, pretrained=True, **kwargs): + model = MobileNetV2( + num_classes, + loss=loss, + width_mult=1.4, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + # init_pretrained_weights(model, model_urls['mobilenetv2_x1_4']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['mobilenetv2_x1_4']) + ) + return model -- Gitee From b97fb1ce9495d2e5c0015a5350b7d385bc82c9e0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:52:02 +0000 Subject: [PATCH 11/31] init --- .../data/__pycache__/__init__.cpython-37.pyc | Bin 0 -> 2539 bytes .../data/__pycache__/datamanager.cpython-37.pyc | Bin 0 -> 16725 bytes .../data/__pycache__/sampler.cpython-37.pyc | Bin 0 -> 8722 bytes .../data/__pycache__/transforms.cpython-37.pyc | Bin 0 -> 11736 bytes 4 files changed, 0 insertions(+), 0 deletions(-) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/datamanager.cpython-37.pyc create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc new file mode 100644 index 0000000000000000000000000000000000000000..04de27608ebbad9017da725d20bea702d90ad355 GIT binary patch literal 2539 zcmb7G&2HO95SE=db~Z_yQ%?mt$SDpKD@}nOlAmpF%S|Unv)#WOV z`x1SQUiuz=1z&s0EA-TFD9Lt^phX3Ez1*FjZ@wAMu>0tdXyNzgAHPi=-*2`4;)l&| z8=D{SS9@5p)~Jo#4mac!FadTB+ev-~`trfi^1sZq301_-RmXkql*Jj2G$Q;S8xNKxpE zyqeKeS6R+Vb)!@&`$~vs^dX{}^!WzLRA-9HWo;3FO#zYZr4#*D!PbU>N?Xv<#$8?u zq5{#eiQCQ)qH#lQvuLd%vQ=wCp2fSW(mEqE;sQ7@rh z*!TP>CMWQ48YGeI4wDcEPn;NVo^$LR`7u5H4@!^|cNqD{j2S${ z!)~0&WSIDL6owuXEn+`9k**)VrhdrOAJQ=P+vGWk(>R6+P=#v*)*Z&OAx#E}A4S7K zBEw)G`8@?Kk#HTTawa_rgC?O!Nf@26QRbi_zfGq-ALmGMuwu$#GGbJP%lX`h3HU>i zB)4Luz(4BCBR_C`Mia7?QyKgF$dv@LBR0u(bn3t-Dg=HZ!vTqAd#hr>FEUB;kR0y> zHUN1gjNHd^T_-q-hi(sKFx@En+l~F@i2{bc{OxYWz$rAF&)nt#)?&htW%dN)-I%l&Ao%SDl=6&l`k zn^~JU5NB%xdeXGzwCO9G9VYUXq>HeZq9ka>&-dbKfbnhoK{dVSC6kPu1|vfM_0S!F(%x2d?m(!YL`Zh$A-V0 z%pEMFy|ts-?C{w=UUqrGcTmgVi>ozyz#`?!hvBa=^8m~5E%O=gdHMTUXRb@t`E~Mh xCyWDSJ64eZ{NSM^@~UTJA$M2g|_uYdQ)n}si8vHzeK?~}sAXZTuOBucCvQ{qZO zkGB%_L@QZOwo>&}obr--x}L^wN*`zq)(3fBTFO(E5F7fvPeYlmaXXCNkF=bH6 zd=*nN>d1>^eH8CQiiCGb9mV??-iMVe-n0CDTuD}Axe@tdCKi*Hsw?cqm1SM-SSnl7 znrhopC285%KQOiJ9h=QHbF6A~TAIDXmUpzaE$cG7e)Zaid8wf5j1yVRR4vtfqAK%} zR8bYpvQ2HXV{1m6$!&!ZYpu;JqhmHzp0la7W%Gb-8D`7Mv!}$|F!`_1v89%wXj@uS zCI)#Xn=0E^O-Nv?3fni0Cz^r>dq=jBs9>V&##61m&6-AA(WsKeE45VnmL$E;q7qrm z*z!ba8j8wV9m_@sY#AaFKY7!5qJoz<0&T;_u;fvuSrTJ9dPgk6jiHZGNiYOWUDjHv zi4i#}0a^-;fF}XErgShsCkTP|NsMnR1dTp%Rg7k*rD6_gkiZL*vw(3oP-e20Y^$aw z>sBzvd}4UT(3J8Xl~v3cDp6ruZmBdP^c)P^j-e|cXoJ5;#D`w9X%sPfLPreKLdyqi zQzbRQI2ufCE64>)lO$;wHs+rXjcqZo)Szc<3kBX_SjLw9l(dAzh5j&WUu}{Cp_)c& zWs>T&g%Vj7x{fZ??v$#my1rU_P^c6cp6^xG@0V7JD{Sc@tKBKG<@I|HE2Z0aYV6MX z+DfrfWrgwza?7gUUS8W+DV1;M85pqgdX24>?v`pGtF7m$QR!$kw!X^l7Awnlz@@NMS}WBaazR#0 zwKBD}ipC1;UZGMeEpMz9D(v1y<=%R=$j~imrBq#BE0pdQSLV?=T4crh#d3{R?-bV7 z`u37!kMwb=2q6ngYek8-gkG+cD#hg*btiaQ#>hdwwLGidD=wGlq4-4+V^F9(%)30R z#UE{e2!+y0;cnq}vC8J&FeI3h<&8@5F3F5>sBSD(Yo*#ot;lY#udmQROVwiKerdT_ z{g|z-)6lQ7jcPH^RtmKOZx}2vR47KrmNu#-K4_(KtyrmS+^dz=%Q?*N0}M+=S}vd} z&EyJOFY^h-l&n`C5>pxjKKOa|;7$?wnBuY+sR9i~6c;XN3$%U%z_&D!bE>pQ;+Gsm-0Xp&Q$pYO!TwKF`){MUq@| zfDOD&OPQ8&KXbw6sN|AMeJS^`bYOJ&qQ=(AH-EloZE3p7)Mrg~-=_VjX|(orO>Seu z^S2q7#k};;WngUbZCU2~inqtOR7|!dX~(wrty>EV*1qy=-Y~ZpbYa4*1xR|uGi>ua zc1xF}jkc~@WUL-{G!vW1<^hxUAw*N&glM|_l(rms+f>Cq1R-hbfMv+{sh9cXtrP^*DS#-n81A}p&LEGX!LX+h1K5$!v54D5`HCB(q zk4h-KSEB8OQICl7J^P;S8@poV#)x zl9}>9C1SB92**f9s4>sQ;$ip{%x&uyf1)?x&{%8^{tnFyxdypgo^#=Lan8!k`!#IY z+*V=6!Zq^D*r%bCb<`CEylD&%)n%nE7*U&T!|t{7J|}8lvl&#GEH

qeredj~)f> z?Wn?u!<6-FhR+2PMDuJ#-I6;x7N^Co-~2FO@>EmcpUs`fgVRJ>vl=zGIAH?l6<=F!-66R<0v@L4JcCivcSD`%O!s+6iZU=dB zP8YWa@r+h>lihT8pgX7}UZ#JKTFu})WJ|llud?67j6dnp`-l?53Qd*|r@h6--E~?J z@EB=guscbCdO1V>G&>n@R&t}xpj+J;beEtrAk^MTbJcaE`?B6qi>7Iqj)bkQZP{d- zoFvw^Njr{{^;^Nt7I{;hv~8Hp9VgZ7D6*5)tcLtV)^su}&Y(MmxgjTIsrr^V4rb;l zN+u{dP00u)qm+zMat_Jg#LNr$`Q?>`9iyc#9B%F{tXIpby}+9?RZUqSZRLr#N$npv znMQ+evyD5bZ2cyZ*ziD{|0Se&CXtOx{%<0l!e0V^te*x+b#{aK(HNaD+9fk2#Y>=nlSn>g3> z8#T`~!FOW!N^5Y%$h(5^LUb>Q;XNWhJUg6kn{e+M z`NT)0F%rA+)hwiQ>#l~Z30)zd%6+Ncj;0!};QB^vM0T*^iC zmuGUdJJDG)4cH7fR2(yaNyp zuYf@})d(NYXBD+=UlO;?c~&D3?}-e!jZU7Nh0%tTJ$&FgY;|<8p?LO`ws=%-c1)9t z2{yV0K#ZRZX*=@tnGwf8seQlbO+G5Qwh^!L*U!7&gEZfNE9P)IL7ATPU2JJUeQG0U^lVgwPa4w8V3fB2BO2IOqEC z)Cn?ZuAk(e9giFZ0H(tBcACJ$NlUH>D=5^4A2>Y`LVFxB5A|3yEA$fD6;eE$q`;HO zio7p?0uJ`eaCd`2+LXxxTiT&2WIa0M!MfzD&LRW|-ifEwoIzw9Zd~urp8T036M77H zqCJ`OQ^VX-fk2D+uvwT7(JincxfVX zaLIsWxY7v6_08gNMe?8~Zsj9qWfy~r%%Nz(WgT=Z!l@xVhJL)ltBcp+OBL~+@Txkd zjb>+O@9K>P&WEZFWTXGtUJwD56hva#1akIUzidw75DwV~j5i?!ifXG(ikpf0A!NZ7 z0%0n^fz5{46Z?xomd~XVyDDjvsDH9RL1|0ILbI>kyn3y1^Xk>TK7E!Ig`wzk0ieoF zxYWEC)VO;9h-eB9G$cmAI5P6?Hp*)hpK*1zZXp^-!$u#{OZU{w7BN`Z&ge)zPx@9A1><0oi<;$ zBEbQH9SWF(#68nt4ni|BA3DqRV@+1$O}vT5WsX(c!#+~T<#Ez{fjF_=!_;d+J!&SU ztFIZ-*K|~OMj}i5^?%{hq(cyW-o=*yog}{2EK)mW10j2n0P2O4J5V*ky)xaQuB60Y zX5ukBsU%*+>drR#Gpe=jXb! zuLis4zCpdmA3TWJneOb1`19HB?Cub7GYP-L_|0}_l=NQ0`Xl@f>?O=!*dr(z?M`+_ zyQjJn-P6k8%K^}hwI}WI-BY_0yQg1G62&)!>5aL0Q{7p}G1eVbGTm_?>$qL;5-48d zj6yiy+3v{Kar4jZX?uobe-Y=ukmj@4i?}kpm-sqvP22A%*;nVN9N*b*&|lz)NoB3G{(CZ zy6*tbnfN5;&MQ-L5qcg%jm{7j_kf$&Sl;f5j}i&@l&=2hnQeRVRPriHk1=HQJ02!N*8z2;ee7s0wPlzG&X5AA z1`Y(=e%hH17d&(Gljj@*bk7Qlc=hq}9U0eJC=jU2`~8Me)bWS%kcCH^cJsL&3Lj@B z*es226<>?sE)7fP%kk$32jDOMmx*rtPZN(*z;F^}AYs1&3X_m&4ksZnn3%t6&Ux5{ON6UlOtjS>BB3%_SuK4=ReVnCJs=a&IdKWD=E!2lCY1M=3Nw& z1NnIWpw+jC{u4?TDfyI=-=X9)N(z)LA<3orq??4-nA_qF^8K5 zE?gbQ(uqrlKVbTF~8tBL#=kUM!P^KDXFUo>R$tJpoON9Vr2VFDf1_AydLO*z-gDAc2d6Oe8Vxc-;-q<4 zTkJ#x9Ghn;cK|>R;2;EnIxPW;zDs2s8GVnQNbAktCejh%4|(gD08S^3D{#8yq-d!- z!=Z}-&cup$p-NTUBZwodfQ+^`RYaG3|D`@Ecnfc;VLM4(Z9^v2=3yjfv|vFQD1mu((Ja)v{`lzExNr3D(S!Q8td^1^lL*wEC+R3{pr61q&PU7sMyq9_GI_&u0_ zGZxOle&P(eJDa&e+>4Z~Qc@$HW0CO$^UtWrUp0=Ts84X!gW^a|hSSn`(DLM4KTg}K z-KbM!P(0f8;o!O9WH?p>8b@*u7kEZcVyMyJCmmqo9Ps8aHv`vy#@E7K%UJBZ#6xBj zHWPoDWCmfhEPoT>G_NFqfZHg(8GOf49+YNNlgRaR5{bB!M364Q;uFX}p4&L*M*IU_ z1E?)>BuFxl&L;3pW_fNvMHD27prv#oC29pZ^wdL>9FLzN`Qzj9Nqi^p9RSrh`jd$t z&K$cH=Vr_s7_1|~32rFMW!Ix#HE)7QBp=cfPso;XI`g;aiIUaTRc>ch z$pRACL#&;KeL&C(y^pxl*i?1hn@XBbgkyH_WAbUG)&Gb1TGVJ}5wXyo_(eFD(f^t6 zAWc;FKa=iL|H#H^9@#jhOhh6Zlg{{kx&cEQfOiY!ZxPPnncTJcKC%(E4d35b2`&tg z;`7@m5%}YK$o~HhvJaa%vOXtawJ=G|J)LfZf$n5HJiyYz4RF7_H@*Rz=mo)g_!Bv= z9?_N?0q^etao4l-A?W|_fVibTZouC>s>M6TQ!+~kVF(|;2jTZrJECdVZU(9*TxFbS zacLhQIF7OroSoQ*Sc*2hRl*ldg!3g8*CX|VO96kH$mJaaPNMjCkRN*_d3Fid0ODGg z@PxqLr7!21SH@4zUa<(w3OVV1r5K4vk9-7^ns9p`k&Hdvq07#kxf_Im3##~p#Hk!;q&u!Cs;*9gH(_{HHy)QI|0x#T;~dIJMfzYIPlW- zY#16^Nmq1jkKbnCcl78!0yGfEM%;$wcf>dmw@H2jVj18cbX=1n76-oqzioijzd1n^ z0(s*%c>S9a7Mj7mi{_4o+3?w1k?CJ|{8mtxidKLn)rr0o13!VaQ0FnXBtcpSe+ zj`#*b`&I{6-T42A@eX;{HT;Ie6%8yZj2pTJWDI8q!4nN0w^m7*JyqSOOLP{F@M^p1 zUa@e}ym4{p>A8^Qg%6TA8-z|cbgp?#_dT5fp`Xnu_wYcG1iHyZ6|6IW*ZM{f1}TPc zB!V#Y>^kAOK+@-)p8?W*25$q19}ELWm%69BlZYNiXKb>L!atSxzU3u z5j_BsJ_RJ5aPG10c=wc&emU@b!XDe@u>v6MC{YH0v&VK3Nl-F8DliEwc!-c*MN)=2 ziaxc1NC1x)WP5VXKKpyrMfsP%@THwrMtY>3_FJ6pj`-~jD5D%#Khr$}Tz!P&>eSn5 zWvn|4DzCrez~yIr4l~Dc7z3_8y-P8SvmVFp%q~Sf#5>9#|2s$3Pq{eug!_B?6~$>% zKF&S)>K&WMIzm|Z6sHML^)s9Xqwd!_%PCw`oiKa;KINCsxUKQlIeI?(-J<6l$-)Wy zIOGz*v@n6rNW^&OQUbz^;y46so429C$Ksp+dl2)<;O#lAFdtm*wU567;PQKcd;(m4 z6J&e@d@j>81aR>_y*{R79m%)Dif8&^#R9}d(NVDYXpxT=A1(5Xi;FRb8Txp!U+X9o zEQ-Dr7JS~kN8_+Z38UmklyD%F{yT!>z-EP>s+3%!4U?XzfT1}Mv|L-8$|XvFxGrTd}K;CDT%I0j;)%(^eaR{psA@+vPH=@5)UVf zVxa#9_|y7F6vw9D76T3N&oE+1h-rX2rht8(gjIUOgME|CJ&6z zs;JNOkoYd2CH_}tz}Vcyy(50{TSQs-0`t2Zfa^}?6LC5HQvz}=f=^RGn&}@^#^c%K N_`UI;#(o+b`)~C|@aO;l literal 0 HcmV?d00001 diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc new file mode 100644 index 0000000000000000000000000000000000000000..f21c4e416378cd36a7fb00554ccfb8d96e3b2ba3 GIT binary patch literal 8722 zcmds7O>-N`d7c>zzJ?#7C|S|2?RA@}N(cuAwOVKG(t1?_0gyzN0vG@kEmRWh05c?p z9AKcHhDd4$RW55Qw^Uq}98x(^u2k8%iexA2`xVf}sDER&ApZ>a?%qYrlC^PxR@$dkb|940%rKwn|rMYTP zYihk%Gp15L=ITvd-s8=LyeFGU+;unAOE=T9PTVzm3(bYzVso*#)LcSY!b(~xD{UD^ z@#gZjVl7yS&rxbGJdZW6A-`y)kx$F~3i3;qfxIE}*R31Yb!+u=wYh4=>PqIO`K^?q z7@PGX`{26;*Btsb+jTm2-?t5;;0?~XbJP`Vy^~>eZ`gB0mle8BUzo1R?%sL-!>o~a zT_!8}jN87=pW0T=Flx5t_<}p_p>VuDGy4`J(N3TF-jH`}S<-g;CO>C~9`E^Cc1prM zF7Mt@7(LH&4xNrk01%PK+EC#ja#8TM?FgY8r zyN7q2^-N)N$8`OvjpfA1hBH$tM^v#fXJ|w(`)1E3i_r7bY+cW_P@xa`lSXfX`xZ(el9MJqPhkF~X@t)p(t)3`L%f(6!}kuwDR~L03;$vM z!0wO(p_xN&#mRO0krVkohK?aLcFJ{DuWmIC^0gAf^Iom`s9Y=++2&)`*eS6>b?9NsEni2=HG^K{E00;}M|-tWz0RsNR(`m*TZS-5 zujMO^a;ctWq zW2wkixly6Mw$NFg?d5BYa$$crUt@dwwY_S+#4s$QSgsd#^W}%7Vh+8dM^<`Nsx(-A zC%?P<+EJ43b7R~rLCO5)Zpn~6VU)#ktyE~xaHdZMSPuH_W?6l&R4CI!=|?5lAYXf& z9m}kj{&XLe@X{#eALh49b+-NmCc&H(_G_hwq%-VL-`}h^%8mU-iEUS_MKZKeFV!BE z3#Ix;Y`03L-(vgqQkE6-jlAp_B48@K#=tiB>t$)Qa-~tK)%N!qG4t7c?#}rp0OXbNJb$CUA=5tcbK!0$O#>^O5rL?_U z-Y!)NCDNu!QVz=XQU-HXhGyF&lKki(k3Qic=ofWpK!$vr@mSawGs((ZEMI&C8BjT@ zjJdCu$9{sM`hH;t#vpUU>1Toc?wz}L*v`;AwH?;5JKetLdPk1!vx1k)vRz>r#+VKO z;C))kw2WUz7i^v0tc`20Wj->_y`fyxfKCbi60;5+*Jk!v#~uj6qmI`bxQ^Ke;9bJZ zSVYcvJQm=!B`li~S0)%UuEk7Y7+oO-{=JP2e_);EJbtv{Mil06K-2F|h%MI@J=ZYy z`>yR1Sv?s#93av@XXXG(bj&tXbIntN9P^0V5e`8~f(|fC7K9ODjB!bBs53RoL?uA; zYz4`G6PRF|`FgpYH4e&+9h%)KG>jKWu}uCbeVDx96D+VS8{@2fHUP&0Ksj^BZbY?p zJK~B-=U`JgN0;#Dn*F1pd1SLAFdFX@MM@%-7zmE+TZZd^S0x?zuMQ_?yhH?}WVA44 zIZquz&hVPq_C0qf>=yL!xTs{*aKW+<&7mtSrz0}4Fj+UDqRqpl8oKa$lkD|63=j4J z$4rX87S$+QR4mTN#XN~#c>+mzBdSpJdZ6b?tDECOQ<%Oj#-$xnb@J2j@Bo*;j>J}) zD#9?$QkyYLvtpks2)%SGj(eO!D%=xR3L#ONBO+4FgsLb}5AHehzF;dSs0;XWpbAi+otGrTfXzmX6p#YAWxPsDw)>m4|}ak%_bF5*v2{r zRZm~RAtoa`r)6nI;vz{M9rmDK8W+t+0hj+>em0Nu?JHFBI8&Q7j?9NYBUdlequ4V)-Wssv-B<*$%yd)caFOqa^C1 z-%%<^;jIqM|CR2@E4p%g4W+XDmU3J9zM?_Olb;~lZVsd*T4?S8nN_YtU5^f%T9 zg-4>BNr$>`yN7(4w7y2k3iT0}(s2`Z4-z0J)E(;#kLcT7Uzlxzo`eFfI4u3pSM*7)@K@5^|@LK(1F^E$+ zWawiV8gF7HKsMeng`0075H-0w}yTbC{LcX(J~=hE}g`Gs&Wl|EeEs?mR@M1 z)Y<(Yg}y!Sbu}w+@?f2ht?m22`&*nOmw;) zh&N37-QnROu>WgR?K(+GMDq{t@lfABLnLmYo!tW`!JDw>0PcPrHJjI>RX`3#4lVe5 zB;;H4YP>)q-$U9=n@>$gu8Yt(#AecM(Yxjnyx8twpD15R|BR9~Qia?P{k2+cZis$jC!!h&qIevEfMahRe$`{tY6Ue>H+k~A#Zg0)(q(Q0AuJap-KvDJDqG~MwV zNXF)`)#ASoq5K0%zDLQo=MyrMB*BmI;1e6BlDd}qM(Vu|o-co-+5=o)mR@OTnm>bO zqA_7VO0+)aSlTKx{0?Xr+4n)WJR({!Az+P}h*%h+5DaSEaNJhF@y!RC=rpRKnMLs$%vnhatk>#iXdY}@XbbnoWa7IvGnx@1}8_fvt0 z1A2(G{*MI#m~(01ai9Ru9`pZX^w@1j%k$0DPq7#gj`Zb!zyqL0X#r9N2p>-|5V`&e zk*i}Ok4Yj2NJ=nz7eIv2;5c@X7!!Fih*`=>%Kr=8wG)+h1rae(b&P_FTM9}w|4gJ2 zS|Bfj2t$S!0AuvNC{iTxVu|EQeU>F{$0{$$Nh4&t7V#R5Lc$x!SK1{)nVVEB@#+?y z!esQ$kC5&iDq86y*hPE>=!jlQiVRczNmi~WxvrDZx}NeVEu>I&_&4bum+m#Q=;V5% zCHXhFe3GE3iP$9t##SWq7#dOIvu{LRT%3)CU!^}fm!uO&IbVN|2caN91Mz_}RL~8= z;8E->2x9(EFG)~2PJn3ivv#0^Xqq53HHgy;${IvONxV-6+C?fz(oCMc8zfP46#1!idT+hv^_qZ9%DbENv8fatCB|3pEx3_zv(eR@rJ~7Wp4QOa6zH+@mBz zNtTk2C?N~-A5uby%{M5KEOZA?86z^dYi&`wb&M7@IT(N^j$;685n{|DX2HhTr$FD>x9jPEgc&>Qf5 zacqFs0(@#>of(8*oUrh)DnWeLgu4cC7M$=^69UmcYHJE@`}0hCW@{;CW7Z?T*Z_g|5s=&5n9UnMx@D_#6_{K#az}Clyp!e9PpGIUnUcH}w zIwNO#&?67!vdkOMX^iR4pX0G&bfU^|p!~tzw;k)vNtg!sZ#m$*JV~U@mehx>pLX$m zZ0OUzX`=m=Gg%tNtXAKfR|Cqg@*^`24ZXhnRQlN{%kY652lIk`Hj+DTPO%oJMLIqH z&ZNSn{G)uiRnOyF8ooht+K=!hN@#_?zKz`NgiCV@%pB^a;3X}ZuW*r`CdLR8C-{hZ zI021JGB;!( zO2-z5Lowtyw%`PUTv9vj3c@&4#E1BGDkaJY7drUdk0Tx|8ebk*HE+xF#vK0L!1Z%9 zn%ran1Yq(rLS)fh%3d)WDK|4v$r5YSPwI!!Ny>fNZ%CHVaR!6|;crpBiQ~$YzOudY M@ydOi^x*Hm0A?!|O#lD@ literal 0 HcmV?d00001 diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc new file mode 100644 index 0000000000000000000000000000000000000000..4fe0c3358c828856ba7b970411f9834e65452b7a GIT binary patch literal 11736 zcmcgyTZ|mpS*}aprsnGLtnKw)a@JXPr#IsskHhBTy^MQ$+MXuko=JC)z3yab*K}2P zm#4a`_f+-RGuxAhc#$Azl|@NFL`aMU4~rB9BqUG}$_nyAL{>cT5Tz6d0kn&GLJ>i* zL<--3s`@hH@e)8#tvX%jQvW&UKmYmu%c-UDaZSSYt*?Dy^YaUm^!JqMT~Rz-#O?k$ z5<@CWhHNOd+*Ha+Q!T5_NI4=?nQBMNQSl!u$M7GqlmM%+1+(^78842_F9ko1z{D`3;uZjFa$d4Lh z$d8Hq3FOC(3FIe4{-iN!Oc}?F>AGqhH)f27>S|3jP8cU&lgp%wGERGkTlI#=W~)h7bUIDTYp_hiYI(Y?v-9T`J~F4JZJUWk zF5{+a@@>;dXBhprjxlV^yO;NIGwRFD2Y7TF@bL zI-b^a469~Ubs{jwbZ)Y?$uR=YG+3KE+m?X`uc3QLOc1ec=a$u~v#QfFEPBZmFEvf? zf~Gyg4vxrWPOUeTs$-a}*>OEgz|%2AqNi^<+a_rBIM8xDuw@Q)ma8#lV|GMhxHHW0 z;7GuNs;yg1lY=Aoj{rRlaiBK>OwH(kLGLpN^iRY1?+${vPgD)3+G(0tLt+wWVR7cc zcL#NxHFeMAmTtQP9*cz$FAgoKm{H!un&Bm?Z0SvtI6}_@wly5vK!X3CRw!5YtHAZbC)SPd0DX1GqxyG2q$!-f1X zw{2EQfbg0{V#P^xT7eL`E~bttlvZ*@R$RMOx|S|v8J^b*Ygcp2*=4qLot0LyEVH(L zy^y=SQerD>tIOF!k)`v?D9x7&xuuQL8Vb&&i+J;lMiuG&b(Vc;y^t*y*;;|+uB@-- zKn&Ci>3k`dEzYrAKC`;9oXcOHV<5otYbCatyOJxRS!r#KI@Rufjjdf`SF(l73aF%) za;v%0buo}jxl*3`x`fWsY&~5ll8dD6Fp)vkcSHmUG3-YC3l%yPQJr=#gcw zX7eRhTuHC49yv=I`++$wWiiO~(rQ){Jzw>rm)>)A|>9IRGOGkgj8f`J$ubG^amj+ z4b73|vzJ$Mm$UgymPWHiRIcTU*(BB~hml<-k|amh(&!Tsf_~A4CZvhSA&CWlv63uz ziKUmXf(C{hG{)K&bD^A|sJM|?0W*l*5c*+YKY#B0IkwW#Zn5v1qwyB0NTEmt20}-+49&KnSA{xo z@1IUey9bL$N~(Usu(mBiPJaTJ+VU#3POC~v#GlkRUB~WtW(DJNxR+1Le$)VmO?t>R z(S0)J#|oxv?V5f(gH?B2(@&I~5=;$;`y+W60lH1KqZ!B`|2aZy?#~o3KBu|jaBCOJ zLbos3R=Y~$>0K%wF5-4)k(g3hhHa)8a#=MLGjcOpQw-IJz}|@&Q6u)6RE`^Q*gXmV zctH95a*69L*A}zauPc%il-oPZ3oEc2NVKh18(PJpxdgJRlS10mH!a)pcCfD)u41Tk zblcvUV|8dL7(Zxd8fMLEL0a5~-Zt56!z8lB+0Dkci^&`#a&O3!)k*iMz6n` zUiGlN(6S4uh4atOG27G)T2ScAkmNIAjUXIk0`jEkq3anY1%|RJciIes3Q-{Y3zEc& zohl446bN2WrQx*RXRc{pcP|L4L^+rdn}u8?FECH%b+gBHmi)f`Zw5iXlGy<=lgE<2Zcc2I{Ao4oA= zA|Qr!pg#j^Ksa%rS!#${e_qtdjT`H^)s&F<6ajEPn=ahAA)3@7pt$~au}Z!KUGxu$ zlQOv7N0D^pu2hk_3R1bNB2~H(q^co7X61Y`;>QAJ`cc6uUvB&32L_gm`VrT(YkUIa z_#`D$lpI6yyAnT+zxU?m8&1=l-`(7rUn}NKZyrKbZE({v<_+D`=Y`-4t31`-@e`Gb z)v~-wjk@l25RlKWo)!uu15#ry?$SG+r zN<6$7#Z!#l4f^dy?d6DXXNYLUD0-bi-@>k!Fv z8MC-YNw}szTuf?yEZACpywkFXdxG&O6?-^X3b81(6uyd8^P}8>_V6RL8Iu~H4l0^; zf23vJsst;n(POaa<a1^x2^siKWtx|!{*0Je% ztWtTUquXIkyi$SFQK|5l=)Jcu_(RC@Gn9~A4ka{7K91x$+%CP9n20D-F;P}Uwttae zUBvC4M{*GU3HS;0$%TgF8x^jzT#f`TGo18sEarqi9td%k>n@;F{|78$O@J*p26V=0_c#M`tqyn`VBu(P`~AoL2qZN9O}-XTuI&yio~M zMZv6@+=S&bz|bx2F8%MU+Z$F?tUpTTo=<8(Qz-1?8HK5-#u~xI>ogcvP-e9(MyAx`KJ{s3w zo%!13AMAca+Z>3C>iV+u~}#Q@H$b%u=-!KjWpRUmIsS-rSJ{gK zjz>Ige-vzq&PaCw$bjV7a4FgarE!39zmF{-MmIL-o${bRX2^rGZtN8GL4ZFf$N0K& zQ9cs3lpmIi=$7(^%ngid+(5ny#RSMtbYmx_UsSpP{BE3{YU=ZnXd~u}dAwg6>Is3r z2eoNE`BU`NG2y!mqqP3PVSS?8Fg1Jb1fo$c982gI81#k9ORU<^TP@ReLu5Zf zL3{#kP{Z7Ix1gE=Eg>vT*l-NNTX=N6fN9_{ESqKw zSO@$=!y6-Yl-wl)S`}?H((@>jMOgs6W%$>TX@G`tmEBh-+u0$ zx9Xo-dhs8>`&CqZ=eOH0Xy5sZ#rnpdzWF=(i(g#qy}p=?1}0!s*w=n6^mXI#{VLmL z)ej!v)F=Vg3q$?Tz9F%GlQ=MjgbeS9l8_Vfl)8KRK3TgT-HC@J&4Pz#@BrkKfZFnw z!cXIwjBp^{mf}9{DW8Mu3@KLg{2W>e8y8N$YFJH9#+5&jb`jqvIm&YG@Nm?AjO5ti zhc-Mcz-q&+Zozke+x!Q3-=)Dw^6ukDjC-hQfBhq9P0Ce_{Utn*7CGQH2+icFHzS_9 zALtt#zq}6Qh0ak>Mkai$tIQB`C!i1DDs^%ASQ|kLUjxoMZpF5?OrE9+rP<^c=^?UhnYWUWV0o9RB&yr(hF*@t zI`$AYIK(w!4F4h8h06_?1T4+~7>{1=9wf^A1-z7uKNuOasDH4j{XA-ale#5SAfZJR zZ8V{BDy`z~UsQV$w|k$lJtcsUjB9Z|1z;5CQUFGA9)+mOBp7|n9}Ucrb+V&%TIwLr zgkjCd`(*n4Hf{-r42X}w5Dy!+9mftZ$Hm?Pf|3Grnq;xxhWA<(XGG9%RaDU-0fjkS z#NkBr2KY*zx=1b|oMP&TdkjT)=0ek346+ZE3&|dB8b;7xI2!cL!f{X>CzuON)Io;E z`&^&Y`Q%`{@YP|)4rV4sd+LdvFiD4$e*k=(6=?%*v#p1M*6c}=? zW(TnioTOB%b|*aeAO;+vU(PbPAf29WS!=eevqi|5$u60?*TEXBntBjIgoDf01EvKv()j}J`qgS7jEE}Jb_=djpC_;skuf=R*7-Z>Q=7xk|9@4K{~ zzLKNP&o2a>_f^Zi^>1Pmq#(fW1~&g|i|+G=7`aPU zy#n)-?0LXTg^WC@67>KxhWt7ypSZC069P&Vii5`BrQ2pJITBd>6$Aj=CWTOG4^-eP z1=dtAuip;xlY{!ez!LsKg*yn;n*MQ8zLgq+p%vS#;SdecL{8x(e?+`T0_RT>aaTvA z*sPE-m>da!h`j#tI5j~SpsiO4hX`PKfs!I6B}z6Zxk||qO6f0A?h+PCqcKIe8_lrHlC6SL>fg2K)s?@2R)D9uyNf7%#WsGC<7(wnbDZbR(i+8Q>c= zQQ=0!JQ3MJ%qH56_1%d0aB1I-Ad_8&yD*KFQ^Y@_JT=6bUYWioiPLaIc1SwvxN-gS zVp7THf&B}}3@Ly^TJ9fFgOY@LcaI$VU)FgzdOB!rR?w@b4-6D!fF5W`#Ey z3-f9hfks#v!dC@E#BoyqG2ZQk4@D?(Kh{m$m1P7sNr>-|ubhA|zDAY^DADI3L10t^ z`@a&Fj#67HCm)r1B-Ymvd105Oj4QxHQCz9|h z;YgBN=bxfPEF&!kr?51?hQyBv;4*v_Xcpjw@Hq)h{8#`pT&E0{4$3My7N8CJk^DL( zWlDt53-M=^eF8~Yggb-vzCrnzc)k|?uN)ErVaEIz?LuEtQ>U;S=yzHkLkyEV%ZN&G zOe#aF9EPNx*_7a`Aj7XvLK~mwDG_CB{U^#M6A!km1ixtB4>qqTa{d->v6&Icj7akE zPn$?gCgPl0`J;gf7tmhUgl~cH)3>(2@mr_o|9Y|hYnflM-~A7GFOQWCcqwngA-VU| zLgK66UaY_UZE+1w7YJ#|ViR$Da;Hc~Z6)|voy=q8Jo%2oAMGl2Im|0}6!&8Y^C@sPY=o_ZyNNAAl(&1q);*Q%Kx4G70?A1u zMvSr76$76fYFDP%I|MTBjZm(JTwVSsS{ReS=@|buT26&6v^!E* zcZ})~x|v4pG{!sXjddmCcz9AkIk}e;eMCYE7GI6g`F8IED`XEua{EoiAtmf)I=aW{ z91L9?KA7+dXj%w74-8L+Yv}Cn>7gC=^q>{_a+N{&n8uLD@#lfg1Aouj4Gu^}V-ZnC z{A8$uen^HIgw6(yLG<8yMi8I=DLQaHtd7nNeDAbv1%2M%-H=_(SikEBzbe}2cR$u= zB)+_&Z{P6gl|IaMun-^R;rl&&T8Bd_VlbM#x+mfSRni)Ck$pjqIvI^F1)}S4wqG72 zD}e$HzpA_vDF!wUeF%Ur^i0GJd?j_xm%T%<{+{+cEjoP>^5Vxq-wkY+-6uCmUMqv~ zrk=MQAh`P?9(phQ&!7^v28H;16|1uQnAjI|WFbUoupwNQN~Ly>eTcqqgcVJkcy;eK zI8np9yW=Pq&L<;l97l8_Q=!AR-T8i6Fd;R__?&n~OL$);5KU(8AXvenNb92NmsWE_roA zvV67CF?XN%5RFZQtAGj=)}h!W#)3hA>f9V#c$u0U+y=8G<}bXuhtFd?1OLhX;OW%G z92&gHhXIz|M+fXa06&7;$>Y9;FHKq`##NKnhx-v)Y+t)dM>rw?<;MuFbi}5;9r*WE ztOh456!r&wox;hy<7DUYfr(CMibB!yQF9FAjq;NT~E zyV#Fy;d8^f8&n3Lu8j5Hfw=m^IrEUP`%{Od`V@&gNtARv0TTwMK5hl({#ay8o`6|H z$FgGx{^1-{o)$sCv7m+WxSUX@@SH$BrHVYG(lg!>`6(rV_EQu#R1{E&pgsQ#dP+Vl zV&`;nTkDX2k0V;zb)1|vPWw9WX~-J%M1K`<5HI|$ppd_u@g@P$yS z&j~kJ2jXwQZd#&%`?NNG%-OsNs^VlfSPBX|1#3iO9AcdCLJ4I2#Ph*P^NaJyi7zK2 Zh>}rYbWGeVI`#C_g{coteOf{OKLI$&eFy*m literal 0 HcmV?d00001 -- Gitee From c5be0845bd703e04a44fa98dbf3f94b7b783c13d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 09:55:52 +0000 Subject: [PATCH 12/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/torchreid?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../OSNet/torchreid/__init__.py | 56 -- .../OSNet/torchreid/data/__init__.py | 54 -- .../data/__pycache__/__init__.cpython-37.pyc | Bin 2539 -> 0 bytes .../__pycache__/datamanager.cpython-37.pyc | Bin 16725 -> 0 bytes .../data/__pycache__/sampler.cpython-37.pyc | Bin 8722 -> 0 bytes .../__pycache__/transforms.cpython-37.pyc | Bin 11736 -> 0 bytes .../OSNet/torchreid/data/datamanager.py | 609 ------------------ .../OSNet/torchreid/data/sampler.py | 292 --------- .../OSNet/torchreid/data/transforms.py | 373 ----------- .../OSNet/torchreid/engine/__init__.py | 52 -- .../OSNet/torchreid/engine/engine.py | 547 ---------------- .../OSNet/torchreid/losses/__init__.py | 68 -- .../torchreid/losses/cross_entropy_loss.py | 100 --- .../losses/hard_mine_triplet_loss.py | 95 --- .../OSNet/torchreid/metrics/__init__.py | 52 -- .../OSNet/torchreid/metrics/accuracy.py | 84 --- .../OSNet/torchreid/metrics/distance.py | 127 ---- .../OSNet/torchreid/metrics/rank.py | 254 -------- .../OSNet/torchreid/models/densenet.py | 425 ------------ .../OSNet/torchreid/models/hacnn.py | 461 ------------- .../torchreid/models/inceptionresnetv2.py | 406 ------------ .../OSNet/torchreid/models/inceptionv4.py | 428 ------------ .../OSNet/torchreid/models/mlfn.py | 316 --------- .../OSNet/torchreid/models/mobilenetv2.py | 321 --------- 24 files changed, 5120 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/datamanager.cpython-37.pyc delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py delete mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py deleted file mode 100644 index 40669b944c..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py +++ /dev/null @@ -1,56 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from torchreid import data, optim, utils, engine, losses, models, metrics - -__version__ = '1.4.0' -__author__ = 'Kaiyang Zhou' -__homepage__ = 'https://kaiyangzhou.github.io/' -__description__ = 'Deep learning person re-identification in PyTorch' -__url__ = 'https://github.com/KaiyangZhou/deep-person-reid' diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py deleted file mode 100644 index 577ed45750..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py +++ /dev/null @@ -1,54 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .datasets import ( - Dataset, ImageDataset, VideoDataset, register_image_dataset, - register_video_dataset -) -from .datamanager import ImageDataManager, VideoDataManager diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/__init__.cpython-37.pyc deleted file mode 100644 index 04de27608ebbad9017da725d20bea702d90ad355..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 2539 zcmb7G&2HO95SE=db~Z_yQ%?mt$SDpKD@}nOlAmpF%S|Unv)#WOV z`x1SQUiuz=1z&s0EA-TFD9Lt^phX3Ez1*FjZ@wAMu>0tdXyNzgAHPi=-*2`4;)l&| z8=D{SS9@5p)~Jo#4mac!FadTB+ev-~`trfi^1sZq301_-RmXkql*Jj2G$Q;S8xNKxpE zyqeKeS6R+Vb)!@&`$~vs^dX{}^!WzLRA-9HWo;3FO#zYZr4#*D!PbU>N?Xv<#$8?u zq5{#eiQCQ)qH#lQvuLd%vQ=wCp2fSW(mEqE;sQ7@rh z*!TP>CMWQ48YGeI4wDcEPn;NVo^$LR`7u5H4@!^|cNqD{j2S${ z!)~0&WSIDL6owuXEn+`9k**)VrhdrOAJQ=P+vGWk(>R6+P=#v*)*Z&OAx#E}A4S7K zBEw)G`8@?Kk#HTTawa_rgC?O!Nf@26QRbi_zfGq-ALmGMuwu$#GGbJP%lX`h3HU>i zB)4Luz(4BCBR_C`Mia7?QyKgF$dv@LBR0u(bn3t-Dg=HZ!vTqAd#hr>FEUB;kR0y> zHUN1gjNHd^T_-q-hi(sKFx@En+l~F@i2{bc{OxYWz$rAF&)nt#)?&htW%dN)-I%l&Ao%SDl=6&l`k zn^~JU5NB%xdeXGzwCO9G9VYUXq>HeZq9ka>&-dbKfbnhoK{dVSC6kPu1|vfM_0S!F(%x2d?m(!YL`Zh$A-V0 z%pEMFy|ts-?C{w=UUqrGcTmgVi>ozyz#`?!hvBa=^8m~5E%O=gdHMTUXRb@t`E~Mh xCyWDSJ64eZ{NSM^@~UTJA$M2g|_uYdQ)n}si8vHzeK?~}sAXZTuOBucCvQ{qZO zkGB%_L@QZOwo>&}obr--x}L^wN*`zq)(3fBTFO(E5F7fvPeYlmaXXCNkF=bH6 zd=*nN>d1>^eH8CQiiCGb9mV??-iMVe-n0CDTuD}Axe@tdCKi*Hsw?cqm1SM-SSnl7 znrhopC285%KQOiJ9h=QHbF6A~TAIDXmUpzaE$cG7e)Zaid8wf5j1yVRR4vtfqAK%} zR8bYpvQ2HXV{1m6$!&!ZYpu;JqhmHzp0la7W%Gb-8D`7Mv!}$|F!`_1v89%wXj@uS zCI)#Xn=0E^O-Nv?3fni0Cz^r>dq=jBs9>V&##61m&6-AA(WsKeE45VnmL$E;q7qrm z*z!ba8j8wV9m_@sY#AaFKY7!5qJoz<0&T;_u;fvuSrTJ9dPgk6jiHZGNiYOWUDjHv zi4i#}0a^-;fF}XErgShsCkTP|NsMnR1dTp%Rg7k*rD6_gkiZL*vw(3oP-e20Y^$aw z>sBzvd}4UT(3J8Xl~v3cDp6ruZmBdP^c)P^j-e|cXoJ5;#D`w9X%sPfLPreKLdyqi zQzbRQI2ufCE64>)lO$;wHs+rXjcqZo)Szc<3kBX_SjLw9l(dAzh5j&WUu}{Cp_)c& zWs>T&g%Vj7x{fZ??v$#my1rU_P^c6cp6^xG@0V7JD{Sc@tKBKG<@I|HE2Z0aYV6MX z+DfrfWrgwza?7gUUS8W+DV1;M85pqgdX24>?v`pGtF7m$QR!$kw!X^l7Awnlz@@NMS}WBaazR#0 zwKBD}ipC1;UZGMeEpMz9D(v1y<=%R=$j~imrBq#BE0pdQSLV?=T4crh#d3{R?-bV7 z`u37!kMwb=2q6ngYek8-gkG+cD#hg*btiaQ#>hdwwLGidD=wGlq4-4+V^F9(%)30R z#UE{e2!+y0;cnq}vC8J&FeI3h<&8@5F3F5>sBSD(Yo*#ot;lY#udmQROVwiKerdT_ z{g|z-)6lQ7jcPH^RtmKOZx}2vR47KrmNu#-K4_(KtyrmS+^dz=%Q?*N0}M+=S}vd} z&EyJOFY^h-l&n`C5>pxjKKOa|;7$?wnBuY+sR9i~6c;XN3$%U%z_&D!bE>pQ;+Gsm-0Xp&Q$pYO!TwKF`){MUq@| zfDOD&OPQ8&KXbw6sN|AMeJS^`bYOJ&qQ=(AH-EloZE3p7)Mrg~-=_VjX|(orO>Seu z^S2q7#k};;WngUbZCU2~inqtOR7|!dX~(wrty>EV*1qy=-Y~ZpbYa4*1xR|uGi>ua zc1xF}jkc~@WUL-{G!vW1<^hxUAw*N&glM|_l(rms+f>Cq1R-hbfMv+{sh9cXtrP^*DS#-n81A}p&LEGX!LX+h1K5$!v54D5`HCB(q zk4h-KSEB8OQICl7J^P;S8@poV#)x zl9}>9C1SB92**f9s4>sQ;$ip{%x&uyf1)?x&{%8^{tnFyxdypgo^#=Lan8!k`!#IY z+*V=6!Zq^D*r%bCb<`CEylD&%)n%nE7*U&T!|t{7J|}8lvl&#GEH

qeredj~)f> z?Wn?u!<6-FhR+2PMDuJ#-I6;x7N^Co-~2FO@>EmcpUs`fgVRJ>vl=zGIAH?l6<=F!-66R<0v@L4JcCivcSD`%O!s+6iZU=dB zP8YWa@r+h>lihT8pgX7}UZ#JKTFu})WJ|llud?67j6dnp`-l?53Qd*|r@h6--E~?J z@EB=guscbCdO1V>G&>n@R&t}xpj+J;beEtrAk^MTbJcaE`?B6qi>7Iqj)bkQZP{d- zoFvw^Njr{{^;^Nt7I{;hv~8Hp9VgZ7D6*5)tcLtV)^su}&Y(MmxgjTIsrr^V4rb;l zN+u{dP00u)qm+zMat_Jg#LNr$`Q?>`9iyc#9B%F{tXIpby}+9?RZUqSZRLr#N$npv znMQ+evyD5bZ2cyZ*ziD{|0Se&CXtOx{%<0l!e0V^te*x+b#{aK(HNaD+9fk2#Y>=nlSn>g3> z8#T`~!FOW!N^5Y%$h(5^LUb>Q;XNWhJUg6kn{e+M z`NT)0F%rA+)hwiQ>#l~Z30)zd%6+Ncj;0!};QB^vM0T*^iC zmuGUdJJDG)4cH7fR2(yaNyp zuYf@})d(NYXBD+=UlO;?c~&D3?}-e!jZU7Nh0%tTJ$&FgY;|<8p?LO`ws=%-c1)9t z2{yV0K#ZRZX*=@tnGwf8seQlbO+G5Qwh^!L*U!7&gEZfNE9P)IL7ATPU2JJUeQG0U^lVgwPa4w8V3fB2BO2IOqEC z)Cn?ZuAk(e9giFZ0H(tBcACJ$NlUH>D=5^4A2>Y`LVFxB5A|3yEA$fD6;eE$q`;HO zio7p?0uJ`eaCd`2+LXxxTiT&2WIa0M!MfzD&LRW|-ifEwoIzw9Zd~urp8T036M77H zqCJ`OQ^VX-fk2D+uvwT7(JincxfVX zaLIsWxY7v6_08gNMe?8~Zsj9qWfy~r%%Nz(WgT=Z!l@xVhJL)ltBcp+OBL~+@Txkd zjb>+O@9K>P&WEZFWTXGtUJwD56hva#1akIUzidw75DwV~j5i?!ifXG(ikpf0A!NZ7 z0%0n^fz5{46Z?xomd~XVyDDjvsDH9RL1|0ILbI>kyn3y1^Xk>TK7E!Ig`wzk0ieoF zxYWEC)VO;9h-eB9G$cmAI5P6?Hp*)hpK*1zZXp^-!$u#{OZU{w7BN`Z&ge)zPx@9A1><0oi<;$ zBEbQH9SWF(#68nt4ni|BA3DqRV@+1$O}vT5WsX(c!#+~T<#Ez{fjF_=!_;d+J!&SU ztFIZ-*K|~OMj}i5^?%{hq(cyW-o=*yog}{2EK)mW10j2n0P2O4J5V*ky)xaQuB60Y zX5ukBsU%*+>drR#Gpe=jXb! zuLis4zCpdmA3TWJneOb1`19HB?Cub7GYP-L_|0}_l=NQ0`Xl@f>?O=!*dr(z?M`+_ zyQjJn-P6k8%K^}hwI}WI-BY_0yQg1G62&)!>5aL0Q{7p}G1eVbGTm_?>$qL;5-48d zj6yiy+3v{Kar4jZX?uobe-Y=ukmj@4i?}kpm-sqvP22A%*;nVN9N*b*&|lz)NoB3G{(CZ zy6*tbnfN5;&MQ-L5qcg%jm{7j_kf$&Sl;f5j}i&@l&=2hnQeRVRPriHk1=HQJ02!N*8z2;ee7s0wPlzG&X5AA z1`Y(=e%hH17d&(Gljj@*bk7Qlc=hq}9U0eJC=jU2`~8Me)bWS%kcCH^cJsL&3Lj@B z*es226<>?sE)7fP%kk$32jDOMmx*rtPZN(*z;F^}AYs1&3X_m&4ksZnn3%t6&Ux5{ON6UlOtjS>BB3%_SuK4=ReVnCJs=a&IdKWD=E!2lCY1M=3Nw& z1NnIWpw+jC{u4?TDfyI=-=X9)N(z)LA<3orq??4-nA_qF^8K5 zE?gbQ(uqrlKVbTF~8tBL#=kUM!P^KDXFUo>R$tJpoON9Vr2VFDf1_AydLO*z-gDAc2d6Oe8Vxc-;-q<4 zTkJ#x9Ghn;cK|>R;2;EnIxPW;zDs2s8GVnQNbAktCejh%4|(gD08S^3D{#8yq-d!- z!=Z}-&cup$p-NTUBZwodfQ+^`RYaG3|D`@Ecnfc;VLM4(Z9^v2=3yjfv|vFQD1mu((Ja)v{`lzExNr3D(S!Q8td^1^lL*wEC+R3{pr61q&PU7sMyq9_GI_&u0_ zGZxOle&P(eJDa&e+>4Z~Qc@$HW0CO$^UtWrUp0=Ts84X!gW^a|hSSn`(DLM4KTg}K z-KbM!P(0f8;o!O9WH?p>8b@*u7kEZcVyMyJCmmqo9Ps8aHv`vy#@E7K%UJBZ#6xBj zHWPoDWCmfhEPoT>G_NFqfZHg(8GOf49+YNNlgRaR5{bB!M364Q;uFX}p4&L*M*IU_ z1E?)>BuFxl&L;3pW_fNvMHD27prv#oC29pZ^wdL>9FLzN`Qzj9Nqi^p9RSrh`jd$t z&K$cH=Vr_s7_1|~32rFMW!Ix#HE)7QBp=cfPso;XI`g;aiIUaTRc>ch z$pRACL#&;KeL&C(y^pxl*i?1hn@XBbgkyH_WAbUG)&Gb1TGVJ}5wXyo_(eFD(f^t6 zAWc;FKa=iL|H#H^9@#jhOhh6Zlg{{kx&cEQfOiY!ZxPPnncTJcKC%(E4d35b2`&tg z;`7@m5%}YK$o~HhvJaa%vOXtawJ=G|J)LfZf$n5HJiyYz4RF7_H@*Rz=mo)g_!Bv= z9?_N?0q^etao4l-A?W|_fVibTZouC>s>M6TQ!+~kVF(|;2jTZrJECdVZU(9*TxFbS zacLhQIF7OroSoQ*Sc*2hRl*ldg!3g8*CX|VO96kH$mJaaPNMjCkRN*_d3Fid0ODGg z@PxqLr7!21SH@4zUa<(w3OVV1r5K4vk9-7^ns9p`k&Hdvq07#kxf_Im3##~p#Hk!;q&u!Cs;*9gH(_{HHy)QI|0x#T;~dIJMfzYIPlW- zY#16^Nmq1jkKbnCcl78!0yGfEM%;$wcf>dmw@H2jVj18cbX=1n76-oqzioijzd1n^ z0(s*%c>S9a7Mj7mi{_4o+3?w1k?CJ|{8mtxidKLn)rr0o13!VaQ0FnXBtcpSe+ zj`#*b`&I{6-T42A@eX;{HT;Ie6%8yZj2pTJWDI8q!4nN0w^m7*JyqSOOLP{F@M^p1 zUa@e}ym4{p>A8^Qg%6TA8-z|cbgp?#_dT5fp`Xnu_wYcG1iHyZ6|6IW*ZM{f1}TPc zB!V#Y>^kAOK+@-)p8?W*25$q19}ELWm%69BlZYNiXKb>L!atSxzU3u z5j_BsJ_RJ5aPG10c=wc&emU@b!XDe@u>v6MC{YH0v&VK3Nl-F8DliEwc!-c*MN)=2 ziaxc1NC1x)WP5VXKKpyrMfsP%@THwrMtY>3_FJ6pj`-~jD5D%#Khr$}Tz!P&>eSn5 zWvn|4DzCrez~yIr4l~Dc7z3_8y-P8SvmVFp%q~Sf#5>9#|2s$3Pq{eug!_B?6~$>% zKF&S)>K&WMIzm|Z6sHML^)s9Xqwd!_%PCw`oiKa;KINCsxUKQlIeI?(-J<6l$-)Wy zIOGz*v@n6rNW^&OQUbz^;y46so429C$Ksp+dl2)<;O#lAFdtm*wU567;PQKcd;(m4 z6J&e@d@j>81aR>_y*{R79m%)Dif8&^#R9}d(NVDYXpxT=A1(5Xi;FRb8Txp!U+X9o zEQ-Dr7JS~kN8_+Z38UmklyD%F{yT!>z-EP>s+3%!4U?XzfT1}Mv|L-8$|XvFxGrTd}K;CDT%I0j;)%(^eaR{psA@+vPH=@5)UVf zVxa#9_|y7F6vw9D76T3N&oE+1h-rX2rht8(gjIUOgME|CJ&6z zs;JNOkoYd2CH_}tz}Vcyy(50{TSQs-0`t2Zfa^}?6LC5HQvz}=f=^RGn&}@^#^c%K N_`UI;#(o+b`)~C|@aO;l diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/sampler.cpython-37.pyc deleted file mode 100644 index f21c4e416378cd36a7fb00554ccfb8d96e3b2ba3..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 8722 zcmds7O>-N`d7c>zzJ?#7C|S|2?RA@}N(cuAwOVKG(t1?_0gyzN0vG@kEmRWh05c?p z9AKcHhDd4$RW55Qw^Uq}98x(^u2k8%iexA2`xVf}sDER&ApZ>a?%qYrlC^PxR@$dkb|940%rKwn|rMYTP zYihk%Gp15L=ITvd-s8=LyeFGU+;unAOE=T9PTVzm3(bYzVso*#)LcSY!b(~xD{UD^ z@#gZjVl7yS&rxbGJdZW6A-`y)kx$F~3i3;qfxIE}*R31Yb!+u=wYh4=>PqIO`K^?q z7@PGX`{26;*Btsb+jTm2-?t5;;0?~XbJP`Vy^~>eZ`gB0mle8BUzo1R?%sL-!>o~a zT_!8}jN87=pW0T=Flx5t_<}p_p>VuDGy4`J(N3TF-jH`}S<-g;CO>C~9`E^Cc1prM zF7Mt@7(LH&4xNrk01%PK+EC#ja#8TM?FgY8r zyN7q2^-N)N$8`OvjpfA1hBH$tM^v#fXJ|w(`)1E3i_r7bY+cW_P@xa`lSXfX`xZ(el9MJqPhkF~X@t)p(t)3`L%f(6!}kuwDR~L03;$vM z!0wO(p_xN&#mRO0krVkohK?aLcFJ{DuWmIC^0gAf^Iom`s9Y=++2&)`*eS6>b?9NsEni2=HG^K{E00;}M|-tWz0RsNR(`m*TZS-5 zujMO^a;ctWq zW2wkixly6Mw$NFg?d5BYa$$crUt@dwwY_S+#4s$QSgsd#^W}%7Vh+8dM^<`Nsx(-A zC%?P<+EJ43b7R~rLCO5)Zpn~6VU)#ktyE~xaHdZMSPuH_W?6l&R4CI!=|?5lAYXf& z9m}kj{&XLe@X{#eALh49b+-NmCc&H(_G_hwq%-VL-`}h^%8mU-iEUS_MKZKeFV!BE z3#Ix;Y`03L-(vgqQkE6-jlAp_B48@K#=tiB>t$)Qa-~tK)%N!qG4t7c?#}rp0OXbNJb$CUA=5tcbK!0$O#>^O5rL?_U z-Y!)NCDNu!QVz=XQU-HXhGyF&lKki(k3Qic=ofWpK!$vr@mSawGs((ZEMI&C8BjT@ zjJdCu$9{sM`hH;t#vpUU>1Toc?wz}L*v`;AwH?;5JKetLdPk1!vx1k)vRz>r#+VKO z;C))kw2WUz7i^v0tc`20Wj->_y`fyxfKCbi60;5+*Jk!v#~uj6qmI`bxQ^Ke;9bJZ zSVYcvJQm=!B`li~S0)%UuEk7Y7+oO-{=JP2e_);EJbtv{Mil06K-2F|h%MI@J=ZYy z`>yR1Sv?s#93av@XXXG(bj&tXbIntN9P^0V5e`8~f(|fC7K9ODjB!bBs53RoL?uA; zYz4`G6PRF|`FgpYH4e&+9h%)KG>jKWu}uCbeVDx96D+VS8{@2fHUP&0Ksj^BZbY?p zJK~B-=U`JgN0;#Dn*F1pd1SLAFdFX@MM@%-7zmE+TZZd^S0x?zuMQ_?yhH?}WVA44 zIZquz&hVPq_C0qf>=yL!xTs{*aKW+<&7mtSrz0}4Fj+UDqRqpl8oKa$lkD|63=j4J z$4rX87S$+QR4mTN#XN~#c>+mzBdSpJdZ6b?tDECOQ<%Oj#-$xnb@J2j@Bo*;j>J}) zD#9?$QkyYLvtpks2)%SGj(eO!D%=xR3L#ONBO+4FgsLb}5AHehzF;dSs0;XWpbAi+otGrTfXzmX6p#YAWxPsDw)>m4|}ak%_bF5*v2{r zRZm~RAtoa`r)6nI;vz{M9rmDK8W+t+0hj+>em0Nu?JHFBI8&Q7j?9NYBUdlequ4V)-Wssv-B<*$%yd)caFOqa^C1 z-%%<^;jIqM|CR2@E4p%g4W+XDmU3J9zM?_Olb;~lZVsd*T4?S8nN_YtU5^f%T9 zg-4>BNr$>`yN7(4w7y2k3iT0}(s2`Z4-z0J)E(;#kLcT7Uzlxzo`eFfI4u3pSM*7)@K@5^|@LK(1F^E$+ zWawiV8gF7HKsMeng`0075H-0w}yTbC{LcX(J~=hE}g`Gs&Wl|EeEs?mR@M1 z)Y<(Yg}y!Sbu}w+@?f2ht?m22`&*nOmw;) zh&N37-QnROu>WgR?K(+GMDq{t@lfABLnLmYo!tW`!JDw>0PcPrHJjI>RX`3#4lVe5 zB;;H4YP>)q-$U9=n@>$gu8Yt(#AecM(Yxjnyx8twpD15R|BR9~Qia?P{k2+cZis$jC!!h&qIevEfMahRe$`{tY6Ue>H+k~A#Zg0)(q(Q0AuJap-KvDJDqG~MwV zNXF)`)#ASoq5K0%zDLQo=MyrMB*BmI;1e6BlDd}qM(Vu|o-co-+5=o)mR@OTnm>bO zqA_7VO0+)aSlTKx{0?Xr+4n)WJR({!Az+P}h*%h+5DaSEaNJhF@y!RC=rpRKnMLs$%vnhatk>#iXdY}@XbbnoWa7IvGnx@1}8_fvt0 z1A2(G{*MI#m~(01ai9Ru9`pZX^w@1j%k$0DPq7#gj`Zb!zyqL0X#r9N2p>-|5V`&e zk*i}Ok4Yj2NJ=nz7eIv2;5c@X7!!Fih*`=>%Kr=8wG)+h1rae(b&P_FTM9}w|4gJ2 zS|Bfj2t$S!0AuvNC{iTxVu|EQeU>F{$0{$$Nh4&t7V#R5Lc$x!SK1{)nVVEB@#+?y z!esQ$kC5&iDq86y*hPE>=!jlQiVRczNmi~WxvrDZx}NeVEu>I&_&4bum+m#Q=;V5% zCHXhFe3GE3iP$9t##SWq7#dOIvu{LRT%3)CU!^}fm!uO&IbVN|2caN91Mz_}RL~8= z;8E->2x9(EFG)~2PJn3ivv#0^Xqq53HHgy;${IvONxV-6+C?fz(oCMc8zfP46#1!idT+hv^_qZ9%DbENv8fatCB|3pEx3_zv(eR@rJ~7Wp4QOa6zH+@mBz zNtTk2C?N~-A5uby%{M5KEOZA?86z^dYi&`wb&M7@IT(N^j$;685n{|DX2HhTr$FD>x9jPEgc&>Qf5 zacqFs0(@#>of(8*oUrh)DnWeLgu4cC7M$=^69UmcYHJE@`}0hCW@{;CW7Z?T*Z_g|5s=&5n9UnMx@D_#6_{K#az}Clyp!e9PpGIUnUcH}w zIwNO#&?67!vdkOMX^iR4pX0G&bfU^|p!~tzw;k)vNtg!sZ#m$*JV~U@mehx>pLX$m zZ0OUzX`=m=Gg%tNtXAKfR|Cqg@*^`24ZXhnRQlN{%kY652lIk`Hj+DTPO%oJMLIqH z&ZNSn{G)uiRnOyF8ooht+K=!hN@#_?zKz`NgiCV@%pB^a;3X}ZuW*r`CdLR8C-{hZ zI021JGB;!( zO2-z5Lowtyw%`PUTv9vj3c@&4#E1BGDkaJY7drUdk0Tx|8ebk*HE+xF#vK0L!1Z%9 zn%ran1Yq(rLS)fh%3d)WDK|4v$r5YSPwI!!Ny>fNZ%CHVaR!6|;crpBiQ~$YzOudY M@ydOi^x*Hm0A?!|O#lD@ diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__pycache__/transforms.cpython-37.pyc deleted file mode 100644 index 4fe0c3358c828856ba7b970411f9834e65452b7a..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 11736 zcmcgyTZ|mpS*}aprsnGLtnKw)a@JXPr#IsskHhBTy^MQ$+MXuko=JC)z3yab*K}2P zm#4a`_f+-RGuxAhc#$Azl|@NFL`aMU4~rB9BqUG}$_nyAL{>cT5Tz6d0kn&GLJ>i* zL<--3s`@hH@e)8#tvX%jQvW&UKmYmu%c-UDaZSSYt*?Dy^YaUm^!JqMT~Rz-#O?k$ z5<@CWhHNOd+*Ha+Q!T5_NI4=?nQBMNQSl!u$M7GqlmM%+1+(^78842_F9ko1z{D`3;uZjFa$d4Lh z$d8Hq3FOC(3FIe4{-iN!Oc}?F>AGqhH)f27>S|3jP8cU&lgp%wGERGkTlI#=W~)h7bUIDTYp_hiYI(Y?v-9T`J~F4JZJUWk zF5{+a@@>;dXBhprjxlV^yO;NIGwRFD2Y7TF@bL zI-b^a469~Ubs{jwbZ)Y?$uR=YG+3KE+m?X`uc3QLOc1ec=a$u~v#QfFEPBZmFEvf? zf~Gyg4vxrWPOUeTs$-a}*>OEgz|%2AqNi^<+a_rBIM8xDuw@Q)ma8#lV|GMhxHHW0 z;7GuNs;yg1lY=Aoj{rRlaiBK>OwH(kLGLpN^iRY1?+${vPgD)3+G(0tLt+wWVR7cc zcL#NxHFeMAmTtQP9*cz$FAgoKm{H!un&Bm?Z0SvtI6}_@wly5vK!X3CRw!5YtHAZbC)SPd0DX1GqxyG2q$!-f1X zw{2EQfbg0{V#P^xT7eL`E~bttlvZ*@R$RMOx|S|v8J^b*Ygcp2*=4qLot0LyEVH(L zy^y=SQerD>tIOF!k)`v?D9x7&xuuQL8Vb&&i+J;lMiuG&b(Vc;y^t*y*;;|+uB@-- zKn&Ci>3k`dEzYrAKC`;9oXcOHV<5otYbCatyOJxRS!r#KI@Rufjjdf`SF(l73aF%) za;v%0buo}jxl*3`x`fWsY&~5ll8dD6Fp)vkcSHmUG3-YC3l%yPQJr=#gcw zX7eRhTuHC49yv=I`++$wWiiO~(rQ){Jzw>rm)>)A|>9IRGOGkgj8f`J$ubG^amj+ z4b73|vzJ$Mm$UgymPWHiRIcTU*(BB~hml<-k|amh(&!Tsf_~A4CZvhSA&CWlv63uz ziKUmXf(C{hG{)K&bD^A|sJM|?0W*l*5c*+YKY#B0IkwW#Zn5v1qwyB0NTEmt20}-+49&KnSA{xo z@1IUey9bL$N~(Usu(mBiPJaTJ+VU#3POC~v#GlkRUB~WtW(DJNxR+1Le$)VmO?t>R z(S0)J#|oxv?V5f(gH?B2(@&I~5=;$;`y+W60lH1KqZ!B`|2aZy?#~o3KBu|jaBCOJ zLbos3R=Y~$>0K%wF5-4)k(g3hhHa)8a#=MLGjcOpQw-IJz}|@&Q6u)6RE`^Q*gXmV zctH95a*69L*A}zauPc%il-oPZ3oEc2NVKh18(PJpxdgJRlS10mH!a)pcCfD)u41Tk zblcvUV|8dL7(Zxd8fMLEL0a5~-Zt56!z8lB+0Dkci^&`#a&O3!)k*iMz6n` zUiGlN(6S4uh4atOG27G)T2ScAkmNIAjUXIk0`jEkq3anY1%|RJciIes3Q-{Y3zEc& zohl446bN2WrQx*RXRc{pcP|L4L^+rdn}u8?FECH%b+gBHmi)f`Zw5iXlGy<=lgE<2Zcc2I{Ao4oA= zA|Qr!pg#j^Ksa%rS!#${e_qtdjT`H^)s&F<6ajEPn=ahAA)3@7pt$~au}Z!KUGxu$ zlQOv7N0D^pu2hk_3R1bNB2~H(q^co7X61Y`;>QAJ`cc6uUvB&32L_gm`VrT(YkUIa z_#`D$lpI6yyAnT+zxU?m8&1=l-`(7rUn}NKZyrKbZE({v<_+D`=Y`-4t31`-@e`Gb z)v~-wjk@l25RlKWo)!uu15#ry?$SG+r zN<6$7#Z!#l4f^dy?d6DXXNYLUD0-bi-@>k!Fv z8MC-YNw}szTuf?yEZACpywkFXdxG&O6?-^X3b81(6uyd8^P}8>_V6RL8Iu~H4l0^; zf23vJsst;n(POaa<a1^x2^siKWtx|!{*0Je% ztWtTUquXIkyi$SFQK|5l=)Jcu_(RC@Gn9~A4ka{7K91x$+%CP9n20D-F;P}Uwttae zUBvC4M{*GU3HS;0$%TgF8x^jzT#f`TGo18sEarqi9td%k>n@;F{|78$O@J*p26V=0_c#M`tqyn`VBu(P`~AoL2qZN9O}-XTuI&yio~M zMZv6@+=S&bz|bx2F8%MU+Z$F?tUpTTo=<8(Qz-1?8HK5-#u~xI>ogcvP-e9(MyAx`KJ{s3w zo%!13AMAca+Z>3C>iV+u~}#Q@H$b%u=-!KjWpRUmIsS-rSJ{gK zjz>Ige-vzq&PaCw$bjV7a4FgarE!39zmF{-MmIL-o${bRX2^rGZtN8GL4ZFf$N0K& zQ9cs3lpmIi=$7(^%ngid+(5ny#RSMtbYmx_UsSpP{BE3{YU=ZnXd~u}dAwg6>Is3r z2eoNE`BU`NG2y!mqqP3PVSS?8Fg1Jb1fo$c982gI81#k9ORU<^TP@ReLu5Zf zL3{#kP{Z7Ix1gE=Eg>vT*l-NNTX=N6fN9_{ESqKw zSO@$=!y6-Yl-wl)S`}?H((@>jMOgs6W%$>TX@G`tmEBh-+u0$ zx9Xo-dhs8>`&CqZ=eOH0Xy5sZ#rnpdzWF=(i(g#qy}p=?1}0!s*w=n6^mXI#{VLmL z)ej!v)F=Vg3q$?Tz9F%GlQ=MjgbeS9l8_Vfl)8KRK3TgT-HC@J&4Pz#@BrkKfZFnw z!cXIwjBp^{mf}9{DW8Mu3@KLg{2W>e8y8N$YFJH9#+5&jb`jqvIm&YG@Nm?AjO5ti zhc-Mcz-q&+Zozke+x!Q3-=)Dw^6ukDjC-hQfBhq9P0Ce_{Utn*7CGQH2+icFHzS_9 zALtt#zq}6Qh0ak>Mkai$tIQB`C!i1DDs^%ASQ|kLUjxoMZpF5?OrE9+rP<^c=^?UhnYWUWV0o9RB&yr(hF*@t zI`$AYIK(w!4F4h8h06_?1T4+~7>{1=9wf^A1-z7uKNuOasDH4j{XA-ale#5SAfZJR zZ8V{BDy`z~UsQV$w|k$lJtcsUjB9Z|1z;5CQUFGA9)+mOBp7|n9}Ucrb+V&%TIwLr zgkjCd`(*n4Hf{-r42X}w5Dy!+9mftZ$Hm?Pf|3Grnq;xxhWA<(XGG9%RaDU-0fjkS z#NkBr2KY*zx=1b|oMP&TdkjT)=0ek346+ZE3&|dB8b;7xI2!cL!f{X>CzuON)Io;E z`&^&Y`Q%`{@YP|)4rV4sd+LdvFiD4$e*k=(6=?%*v#p1M*6c}=? zW(TnioTOB%b|*aeAO;+vU(PbPAf29WS!=eevqi|5$u60?*TEXBntBjIgoDf01EvKv()j}J`qgS7jEE}Jb_=djpC_;skuf=R*7-Z>Q=7xk|9@4K{~ zzLKNP&o2a>_f^Zi^>1Pmq#(fW1~&g|i|+G=7`aPU zy#n)-?0LXTg^WC@67>KxhWt7ypSZC069P&Vii5`BrQ2pJITBd>6$Aj=CWTOG4^-eP z1=dtAuip;xlY{!ez!LsKg*yn;n*MQ8zLgq+p%vS#;SdecL{8x(e?+`T0_RT>aaTvA z*sPE-m>da!h`j#tI5j~SpsiO4hX`PKfs!I6B}z6Zxk||qO6f0A?h+PCqcKIe8_lrHlC6SL>fg2K)s?@2R)D9uyNf7%#WsGC<7(wnbDZbR(i+8Q>c= zQQ=0!JQ3MJ%qH56_1%d0aB1I-Ad_8&yD*KFQ^Y@_JT=6bUYWioiPLaIc1SwvxN-gS zVp7THf&B}}3@Ly^TJ9fFgOY@LcaI$VU)FgzdOB!rR?w@b4-6D!fF5W`#Ey z3-f9hfks#v!dC@E#BoyqG2ZQk4@D?(Kh{m$m1P7sNr>-|ubhA|zDAY^DADI3L10t^ z`@a&Fj#67HCm)r1B-Ymvd105Oj4QxHQCz9|h z;YgBN=bxfPEF&!kr?51?hQyBv;4*v_Xcpjw@Hq)h{8#`pT&E0{4$3My7N8CJk^DL( zWlDt53-M=^eF8~Yggb-vzCrnzc)k|?uN)ErVaEIz?LuEtQ>U;S=yzHkLkyEV%ZN&G zOe#aF9EPNx*_7a`Aj7XvLK~mwDG_CB{U^#M6A!km1ixtB4>qqTa{d->v6&Icj7akE zPn$?gCgPl0`J;gf7tmhUgl~cH)3>(2@mr_o|9Y|hYnflM-~A7GFOQWCcqwngA-VU| zLgK66UaY_UZE+1w7YJ#|ViR$Da;Hc~Z6)|voy=q8Jo%2oAMGl2Im|0}6!&8Y^C@sPY=o_ZyNNAAl(&1q);*Q%Kx4G70?A1u zMvSr76$76fYFDP%I|MTBjZm(JTwVSsS{ReS=@|buT26&6v^!E* zcZ})~x|v4pG{!sXjddmCcz9AkIk}e;eMCYE7GI6g`F8IED`XEua{EoiAtmf)I=aW{ z91L9?KA7+dXj%w74-8L+Yv}Cn>7gC=^q>{_a+N{&n8uLD@#lfg1Aouj4Gu^}V-ZnC z{A8$uen^HIgw6(yLG<8yMi8I=DLQaHtd7nNeDAbv1%2M%-H=_(SikEBzbe}2cR$u= zB)+_&Z{P6gl|IaMun-^R;rl&&T8Bd_VlbM#x+mfSRni)Ck$pjqIvI^F1)}S4wqG72 zD}e$HzpA_vDF!wUeF%Ur^i0GJd?j_xm%T%<{+{+cEjoP>^5Vxq-wkY+-6uCmUMqv~ zrk=MQAh`P?9(phQ&!7^v28H;16|1uQnAjI|WFbUoupwNQN~Ly>eTcqqgcVJkcy;eK zI8np9yW=Pq&L<;l97l8_Q=!AR-T8i6Fd;R__?&n~OL$);5KU(8AXvenNb92NmsWE_roA zvV67CF?XN%5RFZQtAGj=)}h!W#)3hA>f9V#c$u0U+y=8G<}bXuhtFd?1OLhX;OW%G z92&gHhXIz|M+fXa06&7;$>Y9;FHKq`##NKnhx-v)Y+t)dM>rw?<;MuFbi}5;9r*WE ztOh456!r&wox;hy<7DUYfr(CMibB!yQF9FAjq;NT~E zyV#Fy;d8^f8&n3Lu8j5Hfw=m^IrEUP`%{Od`V@&gNtARv0TTwMK5hl({#ay8o`6|H z$FgGx{^1-{o)$sCv7m+WxSUX@@SH$BrHVYG(lg!>`6(rV_EQu#R1{E&pgsQ#dP+Vl zV&`;nTkDX2k0V;zb)1|vPWw9WX~-J%M1K`<5HI|$ppd_u@g@P$yS z&j~kJ2jXwQZd#&%`?NNG%-OsNs^VlfSPBX|1#3iO9AcdCLJ4I2#Ph*P^NaJyi7zK2 Zh>}rYbWGeVI`#C_g{coteOf{OKLI$&eFy*m diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py deleted file mode 100644 index 009d03802f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py +++ /dev/null @@ -1,609 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch - -from torchreid.data.sampler import build_train_sampler -from torchreid.data.datasets import init_image_dataset, init_video_dataset -from torchreid.data.transforms import build_transforms - - -class DataManager(object): - r"""Base data manager. - - Args: - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - """ - - def __init__( - self, - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - norm_mean=None, - norm_std=None, - use_gpu=False - ): - self.sources = sources - self.targets = targets - self.height = height - self.width = width - - if self.sources is None: - raise ValueError('sources must not be None') - - if isinstance(self.sources, str): - self.sources = [self.sources] - - if self.targets is None: - self.targets = self.sources - - if isinstance(self.targets, str): - self.targets = [self.targets] - - self.transform_tr, self.transform_te = build_transforms( - self.height, - self.width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std - ) - - self.use_gpu = (torch.cuda.is_available() and use_gpu) - - @property - def num_train_pids(self): - """Returns the number of training person identities.""" - return self._num_train_pids - - @property - def num_train_cams(self): - """Returns the number of training cameras.""" - return self._num_train_cams - - def fetch_test_loaders(self, name): - """Returns query and gallery of a test dataset, each containing - tuples of (img_path(s), pid, camid). - - Args: - name (str): dataset name. - """ - query_loader = self.test_dataset[name]['query'] - gallery_loader = self.test_dataset[name]['gallery'] - return query_loader, gallery_loader - - def preprocess_pil_img(self, img): - """Transforms a PIL image to torch tensor for testing.""" - return self.transform_te(img) - - -class ImageDataManager(DataManager): - r"""Image data manager. - - Args: - root (str): root path to datasets. - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - k_tfm (int): number of times to apply augmentation to an image - independently. If k_tfm > 1, the transform function will be - applied k_tfm times to an image. This variable will only be - useful for training and is currently valid for image datasets only. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - split_id (int, optional): split id (*0-based*). Default is 0. - combineall (bool, optional): combine train, query and gallery in a dataset for - training. Default is False. - load_train_targets (bool, optional): construct train-loader for target datasets. - Default is False. This is useful for domain adaptation research. - batch_size_train (int, optional): number of images in a training batch. Default is 32. - batch_size_test (int, optional): number of images in a test batch. Default is 32. - workers (int, optional): number of workers. Default is 4. - num_instances (int, optional): number of instances per identity in a batch. - Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - train_sampler (str, optional): sampler. Default is RandomSampler. - train_sampler_t (str, optional): sampler for target train loader. Default is RandomSampler. - cuhk03_labeled (bool, optional): use cuhk03 labeled images. - Default is False (defaul is to use detected images). - cuhk03_classic_split (bool, optional): use the classic split in cuhk03. - Default is False. - market1501_500k (bool, optional): add 500K distractors to the gallery - set in market1501. Default is False. - - Examples:: - - datamanager = torchreid.data.ImageDataManager( - root='path/to/reid-data', - sources='market1501', - height=256, - width=128, - batch_size_train=32, - batch_size_test=100 - ) - - # return train loader of source data - train_loader = datamanager.train_loader - - # return test loader of target data - test_loader = datamanager.test_loader - - # return train loader of target data - train_loader_t = datamanager.train_loader_t - """ - data_type = 'image' - - def __init__( - self, - root='', - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - k_tfm=1, - norm_mean=None, - norm_std=None, - use_gpu=True, - split_id=0, - combineall=False, - load_train_targets=False, - batch_size_train=32, - batch_size_test=32, - workers=4, - num_instances=4, - num_cams=1, - num_datasets=1, - train_sampler='RandomSampler', - train_sampler_t='RandomSampler', - cuhk03_labeled=False, - cuhk03_classic_split=False, - market1501_500k=False, - device_num=-1 - ): - - super(ImageDataManager, self).__init__( - sources=sources, - targets=targets, - height=height, - width=width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std, - use_gpu=use_gpu - ) - - print('=> Loading train (source) dataset') - trainset = [] - for name in self.sources: - trainset_ = init_image_dataset( - name, - transform=self.transform_tr, - k_tfm=k_tfm, - mode='train', - combineall=combineall, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - trainset.append(trainset_) - trainset = sum(trainset) - - self._num_train_pids = trainset.num_train_pids - self._num_train_cams = trainset.num_train_cams - - if device_num == -1 or device_num == 1: - self.train_sampler = build_train_sampler( - trainset.train, - train_sampler, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ) - - else: - self.train_sampler = torch.utils.data.distributed.DistributedSampler(trainset.train) - - self.train_loader = torch.utils.data.DataLoader( - trainset, - sampler=self.train_sampler, - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - # pin_memory=self.use_gpu, - pin_memory=True, - drop_last=True - ) - - self.train_loader_t = None - if load_train_targets: - # check if sources and targets are identical - assert len(set(self.sources) & set(self.targets)) == 0, \ - 'sources={} and targets={} must not have overlap'.format(self.sources, self.targets) - - print('=> Loading train (target) dataset') - trainset_t = [] - for name in self.targets: - trainset_t_ = init_image_dataset( - name, - transform=self.transform_tr, - k_tfm=k_tfm, - mode='train', - combineall=False, # only use the training data - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - trainset_t.append(trainset_t_) - trainset_t = sum(trainset_t) - - self.train_loader_t = torch.utils.data.DataLoader( - trainset_t, - sampler=build_train_sampler( - trainset_t.train, - train_sampler_t, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ), - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=True - ) - - print('=> Loading test (target) dataset') - self.test_loader = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - self.test_dataset = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - - for name in self.targets: - # build query loader - queryset = init_image_dataset( - name, - transform=self.transform_te, - mode='query', - combineall=combineall, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - self.test_loader[name]['query'] = torch.utils.data.DataLoader( - queryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=True, - drop_last=False - ) - - # build gallery loader - galleryset = init_image_dataset( - name, - transform=self.transform_te, - mode='gallery', - combineall=combineall, - verbose=False, - root=root, - split_id=split_id, - cuhk03_labeled=cuhk03_labeled, - cuhk03_classic_split=cuhk03_classic_split, - market1501_500k=market1501_500k - ) - self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( - galleryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=True, - drop_last=False - ) - - self.test_dataset[name]['query'] = queryset.query - self.test_dataset[name]['gallery'] = galleryset.gallery - - print('\n') - print(' **************** Summary ****************') - print(' source : {}'.format(self.sources)) - print(' # source datasets : {}'.format(len(self.sources))) - print(' # source ids : {}'.format(self.num_train_pids)) - print(' # source images : {}'.format(len(trainset))) - print(' # source cameras : {}'.format(self.num_train_cams)) - if load_train_targets: - print( - ' # target images : {} (unlabeled)'.format(len(trainset_t)) - ) - print(' target : {}'.format(self.targets)) - print(' *****************************************') - print('\n') - - -class VideoDataManager(DataManager): - r"""Video data manager. - - Args: - root (str): root path to datasets. - sources (str or list): source dataset(s). - targets (str or list, optional): target dataset(s). If not given, - it equals to ``sources``. - height (int, optional): target image height. Default is 256. - width (int, optional): target image width. Default is 128. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). - norm_std (list or None, optional): data std. Default is None (use imagenet std). - use_gpu (bool, optional): use gpu. Default is True. - split_id (int, optional): split id (*0-based*). Default is 0. - combineall (bool, optional): combine train, query and gallery in a dataset for - training. Default is False. - batch_size_train (int, optional): number of tracklets in a training batch. Default is 3. - batch_size_test (int, optional): number of tracklets in a test batch. Default is 3. - workers (int, optional): number of workers. Default is 4. - num_instances (int, optional): number of instances per identity in a batch. - Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - train_sampler (str, optional): sampler. Default is RandomSampler. - seq_len (int, optional): how many images to sample in a tracklet. Default is 15. - sample_method (str, optional): how to sample images in a tracklet. Default is "evenly". - Choices are ["evenly", "random", "all"]. "evenly" and "random" will sample ``seq_len`` - images in a tracklet while "all" samples all images in a tracklet, where the batch size - needs to be set to 1. - - Examples:: - - datamanager = torchreid.data.VideoDataManager( - root='path/to/reid-data', - sources='mars', - height=256, - width=128, - batch_size_train=3, - batch_size_test=3, - seq_len=15, - sample_method='evenly' - ) - - # return train loader of source data - train_loader = datamanager.train_loader - - # return test loader of target data - test_loader = datamanager.test_loader - - .. note:: - The current implementation only supports image-like training. Therefore, each image in a - sampled tracklet will undergo independent transformation functions. To achieve tracklet-aware - training, you need to modify the transformation functions for video reid such that each function - applies the same operation to all images in a tracklet to keep consistency. - """ - data_type = 'video' - - def __init__( - self, - root='', - sources=None, - targets=None, - height=256, - width=128, - transforms='random_flip', - norm_mean=None, - norm_std=None, - use_gpu=True, - split_id=0, - combineall=False, - batch_size_train=3, - batch_size_test=3, - workers=4, - num_instances=4, - num_cams=1, - num_datasets=1, - train_sampler='RandomSampler', - seq_len=15, - sample_method='evenly' - ): - - super(VideoDataManager, self).__init__( - sources=sources, - targets=targets, - height=height, - width=width, - transforms=transforms, - norm_mean=norm_mean, - norm_std=norm_std, - use_gpu=use_gpu - ) - - print('=> Loading train (source) dataset') - trainset = [] - for name in self.sources: - trainset_ = init_video_dataset( - name, - transform=self.transform_tr, - mode='train', - combineall=combineall, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - trainset.append(trainset_) - trainset = sum(trainset) - - self._num_train_pids = trainset.num_train_pids - self._num_train_cams = trainset.num_train_cams - - train_sampler = build_train_sampler( - trainset.train, - train_sampler, - batch_size=batch_size_train, - num_instances=num_instances, - num_cams=num_cams, - num_datasets=num_datasets - ) - - self.train_loader = torch.utils.data.DataLoader( - trainset, - sampler=train_sampler, - batch_size=batch_size_train, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=True - ) - - print('=> Loading test (target) dataset') - self.test_loader = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - self.test_dataset = { - name: { - 'query': None, - 'gallery': None - } - for name in self.targets - } - - for name in self.targets: - # build query loader - queryset = init_video_dataset( - name, - transform=self.transform_te, - mode='query', - combineall=combineall, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - self.test_loader[name]['query'] = torch.utils.data.DataLoader( - queryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=False - ) - - # build gallery loader - galleryset = init_video_dataset( - name, - transform=self.transform_te, - mode='gallery', - combineall=combineall, - verbose=False, - root=root, - split_id=split_id, - seq_len=seq_len, - sample_method=sample_method - ) - self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( - galleryset, - batch_size=batch_size_test, - shuffle=False, - num_workers=workers, - pin_memory=self.use_gpu, - drop_last=False - ) - - self.test_dataset[name]['query'] = queryset.query - self.test_dataset[name]['gallery'] = galleryset.gallery - - print('\n') - print(' **************** Summary ****************') - print(' source : {}'.format(self.sources)) - print(' # source datasets : {}'.format(len(self.sources))) - print(' # source ids : {}'.format(self.num_train_pids)) - print(' # source tracklets : {}'.format(len(trainset))) - print(' # source cameras : {}'.format(self.num_train_cams)) - print(' target : {}'.format(self.targets)) - print(' *****************************************') - print('\n') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py deleted file mode 100644 index daf0d026c3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py +++ /dev/null @@ -1,292 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import copy -import numpy as np -import random -from collections import defaultdict -from torch.utils.data.sampler import Sampler, RandomSampler, SequentialSampler - -AVAI_SAMPLERS = [ - 'RandomIdentitySampler', 'SequentialSampler', 'RandomSampler', - 'RandomDomainSampler', 'RandomDatasetSampler' -] - - -class RandomIdentitySampler(Sampler): - """Randomly samples N identities each with K instances. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - num_instances (int): number of instances per identity in a batch. - """ - - def __init__(self, data_source, batch_size, num_instances): - if batch_size < num_instances: - raise ValueError( - 'batch_size={} must be no less ' - 'than num_instances={}'.format(batch_size, num_instances) - ) - - self.data_source = data_source - self.batch_size = batch_size - self.num_instances = num_instances - self.num_pids_per_batch = self.batch_size // self.num_instances - self.index_dic = defaultdict(list) - for index, items in enumerate(data_source): - pid = items[1] - self.index_dic[pid].append(index) - self.pids = list(self.index_dic.keys()) - assert len(self.pids) >= self.num_pids_per_batch - - # estimate number of examples in an epoch - # TODO: improve precision - self.length = 0 - for pid in self.pids: - idxs = self.index_dic[pid] - num = len(idxs) - if num < self.num_instances: - num = self.num_instances - self.length += num - num % self.num_instances - - def __iter__(self): - batch_idxs_dict = defaultdict(list) - - for pid in self.pids: - idxs = copy.deepcopy(self.index_dic[pid]) - if len(idxs) < self.num_instances: - idxs = np.random.choice( - idxs, size=self.num_instances, replace=True - ) - random.shuffle(idxs) - batch_idxs = [] - for idx in idxs: - batch_idxs.append(idx) - if len(batch_idxs) == self.num_instances: - batch_idxs_dict[pid].append(batch_idxs) - batch_idxs = [] - - avai_pids = copy.deepcopy(self.pids) - final_idxs = [] - - while len(avai_pids) >= self.num_pids_per_batch: - selected_pids = random.sample(avai_pids, self.num_pids_per_batch) - for pid in selected_pids: - batch_idxs = batch_idxs_dict[pid].pop(0) - final_idxs.extend(batch_idxs) - if len(batch_idxs_dict[pid]) == 0: - avai_pids.remove(pid) - - return iter(final_idxs) - - def __len__(self): - return self.length - - -class RandomDomainSampler(Sampler): - """Random domain sampler. - - We consider each camera as a visual domain. - - How does the sampling work: - 1. Randomly sample N cameras (based on the "camid" label). - 2. From each camera, randomly sample K images. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - n_domain (int): number of cameras to sample in a batch. - """ - - def __init__(self, data_source, batch_size, n_domain): - self.data_source = data_source - - # Keep track of image indices for each domain - self.domain_dict = defaultdict(list) - for i, items in enumerate(data_source): - camid = items[2] - self.domain_dict[camid].append(i) - self.domains = list(self.domain_dict.keys()) - - # Make sure each domain can be assigned an equal number of images - if n_domain is None or n_domain <= 0: - n_domain = len(self.domains) - assert batch_size % n_domain == 0 - self.n_img_per_domain = batch_size // n_domain - - self.batch_size = batch_size - self.n_domain = n_domain - self.length = len(list(self.__iter__())) - - def __iter__(self): - domain_dict = copy.deepcopy(self.domain_dict) - final_idxs = [] - stop_sampling = False - - while not stop_sampling: - selected_domains = random.sample(self.domains, self.n_domain) - - for domain in selected_domains: - idxs = domain_dict[domain] - selected_idxs = random.sample(idxs, self.n_img_per_domain) - final_idxs.extend(selected_idxs) - - for idx in selected_idxs: - domain_dict[domain].remove(idx) - - remaining = len(domain_dict[domain]) - if remaining < self.n_img_per_domain: - stop_sampling = True - - return iter(final_idxs) - - def __len__(self): - return self.length - - -class RandomDatasetSampler(Sampler): - """Random dataset sampler. - - How does the sampling work: - 1. Randomly sample N datasets (based on the "dsetid" label). - 2. From each dataset, randomly sample K images. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). - batch_size (int): batch size. - n_dataset (int): number of datasets to sample in a batch. - """ - - def __init__(self, data_source, batch_size, n_dataset): - self.data_source = data_source - - # Keep track of image indices for each dataset - self.dataset_dict = defaultdict(list) - for i, items in enumerate(data_source): - dsetid = items[3] - self.dataset_dict[dsetid].append(i) - self.datasets = list(self.dataset_dict.keys()) - - # Make sure each dataset can be assigned an equal number of images - if n_dataset is None or n_dataset <= 0: - n_dataset = len(self.datasets) - assert batch_size % n_dataset == 0 - self.n_img_per_dset = batch_size // n_dataset - - self.batch_size = batch_size - self.n_dataset = n_dataset - self.length = len(list(self.__iter__())) - - def __iter__(self): - dataset_dict = copy.deepcopy(self.dataset_dict) - final_idxs = [] - stop_sampling = False - - while not stop_sampling: - selected_datasets = random.sample(self.datasets, self.n_dataset) - - for dset in selected_datasets: - idxs = dataset_dict[dset] - selected_idxs = random.sample(idxs, self.n_img_per_dset) - final_idxs.extend(selected_idxs) - - for idx in selected_idxs: - dataset_dict[dset].remove(idx) - - remaining = len(dataset_dict[dset]) - if remaining < self.n_img_per_dset: - stop_sampling = True - - return iter(final_idxs) - - def __len__(self): - return self.length - - -def build_train_sampler( - data_source, - train_sampler, - batch_size=32, - num_instances=4, - num_cams=1, - num_datasets=1, - **kwargs -): - """Builds a training sampler. - - Args: - data_source (list): contains tuples of (img_path(s), pid, camid). - train_sampler (str): sampler name (default: ``RandomSampler``). - batch_size (int, optional): batch size. Default is 32. - num_instances (int, optional): number of instances per identity in a - batch (when using ``RandomIdentitySampler``). Default is 4. - num_cams (int, optional): number of cameras to sample in a batch (when using - ``RandomDomainSampler``). Default is 1. - num_datasets (int, optional): number of datasets to sample in a batch (when - using ``RandomDatasetSampler``). Default is 1. - """ - assert train_sampler in AVAI_SAMPLERS, \ - 'train_sampler must be one of {}, but got {}'.format(AVAI_SAMPLERS, train_sampler) - - if train_sampler == 'RandomIdentitySampler': - sampler = RandomIdentitySampler(data_source, batch_size, num_instances) - - elif train_sampler == 'RandomDomainSampler': - sampler = RandomDomainSampler(data_source, batch_size, num_cams) - - elif train_sampler == 'RandomDatasetSampler': - sampler = RandomDatasetSampler(data_source, batch_size, num_datasets) - - elif train_sampler == 'SequentialSampler': - sampler = SequentialSampler(data_source) - - elif train_sampler == 'RandomSampler': - sampler = RandomSampler(data_source) - - return sampler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py deleted file mode 100644 index 3108b81565..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py +++ /dev/null @@ -1,373 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import math -import random -from collections import deque -import torch -from PIL import Image -from torchvision.transforms import ( - Resize, Compose, ToTensor, Normalize, ColorJitter, RandomHorizontalFlip -) - - -class Random2DTranslation(object): - """Randomly translates the input image with a probability. - - Specifically, given a predefined shape (height, width), the input is first - resized with a factor of 1.125, leading to (height*1.125, width*1.125), then - a random crop is performed. Such operation is done with a probability. - - Args: - height (int): target image height. - width (int): target image width. - p (float, optional): probability that this operation takes place. - Default is 0.5. - interpolation (int, optional): desired interpolation. Default is - ``PIL.Image.BILINEAR`` - """ - - def __init__(self, height, width, p=0.5, interpolation=Image.BILINEAR): - self.height = height - self.width = width - self.p = p - self.interpolation = interpolation - - def __call__(self, img): - if random.uniform(0, 1) > self.p: - return img.resize((self.width, self.height), self.interpolation) - - new_width, new_height = int(round(self.width * 1.125) - ), int(round(self.height * 1.125)) - resized_img = img.resize((new_width, new_height), self.interpolation) - x_maxrange = new_width - self.width - y_maxrange = new_height - self.height - x1 = int(round(random.uniform(0, x_maxrange))) - y1 = int(round(random.uniform(0, y_maxrange))) - croped_img = resized_img.crop( - (x1, y1, x1 + self.width, y1 + self.height) - ) - return croped_img - - -class RandomErasing(object): - """Randomly erases an image patch. - - Origin: ``_ - - Reference: - Zhong et al. Random Erasing Data Augmentation. - - Args: - probability (float, optional): probability that this operation takes place. - Default is 0.5. - sl (float, optional): min erasing area. - sh (float, optional): max erasing area. - r1 (float, optional): min aspect ratio. - mean (list, optional): erasing value. - """ - - def __init__( - self, - probability=0.5, - sl=0.02, - sh=0.4, - r1=0.3, - mean=[0.4914, 0.4822, 0.4465] - ): - self.probability = probability - self.mean = mean - self.sl = sl - self.sh = sh - self.r1 = r1 - - def __call__(self, img): - if random.uniform(0, 1) > self.probability: - return img - - for attempt in range(100): - area = img.size()[1] * img.size()[2] - - target_area = random.uniform(self.sl, self.sh) * area - aspect_ratio = random.uniform(self.r1, 1 / self.r1) - - h = int(round(math.sqrt(target_area * aspect_ratio))) - w = int(round(math.sqrt(target_area / aspect_ratio))) - - if w < img.size()[2] and h < img.size()[1]: - x1 = random.randint(0, img.size()[1] - h) - y1 = random.randint(0, img.size()[2] - w) - if img.size()[0] == 3: - img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] - img[1, x1:x1 + h, y1:y1 + w] = self.mean[1] - img[2, x1:x1 + h, y1:y1 + w] = self.mean[2] - else: - img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] - return img - - return img - - -class ColorAugmentation(object): - """Randomly alters the intensities of RGB channels. - - Reference: - Krizhevsky et al. ImageNet Classification with Deep ConvolutionalNeural - Networks. NIPS 2012. - - Args: - p (float, optional): probability that this operation takes place. - Default is 0.5. - """ - - def __init__(self, p=0.5): - self.p = p - self.eig_vec = torch.Tensor( - [ - [0.4009, 0.7192, -0.5675], - [-0.8140, -0.0045, -0.5808], - [0.4203, -0.6948, -0.5836], - ] - ) - self.eig_val = torch.Tensor([[0.2175, 0.0188, 0.0045]]) - - def _check_input(self, tensor): - assert tensor.dim() == 3 and tensor.size(0) == 3 - - def __call__(self, tensor): - if random.uniform(0, 1) > self.p: - return tensor - alpha = torch.normal(mean=torch.zeros_like(self.eig_val)) * 0.1 - quatity = torch.mm(self.eig_val * alpha, self.eig_vec) - tensor = tensor + quatity.view(3, 1, 1) - return tensor - - -class RandomPatch(object): - """Random patch data augmentation. - - There is a patch pool that stores randomly extracted pathces from person images. - - For each input image, RandomPatch - 1) extracts a random patch and stores the patch in the patch pool; - 2) randomly selects a patch from the patch pool and pastes it on the - input (at random position) to simulate occlusion. - - Reference: - - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. - - Zhou et al. Learning Generalisable Omni-Scale Representations - for Person Re-Identification. TPAMI, 2021. - """ - - def __init__( - self, - prob_happen=0.5, - pool_capacity=50000, - min_sample_size=100, - patch_min_area=0.01, - patch_max_area=0.5, - patch_min_ratio=0.1, - prob_rotate=0.5, - prob_flip_leftright=0.5, - ): - self.prob_happen = prob_happen - - self.patch_min_area = patch_min_area - self.patch_max_area = patch_max_area - self.patch_min_ratio = patch_min_ratio - - self.prob_rotate = prob_rotate - self.prob_flip_leftright = prob_flip_leftright - - self.patchpool = deque(maxlen=pool_capacity) - self.min_sample_size = min_sample_size - - def generate_wh(self, W, H): - area = W * H - for attempt in range(100): - target_area = random.uniform( - self.patch_min_area, self.patch_max_area - ) * area - aspect_ratio = random.uniform( - self.patch_min_ratio, 1. / self.patch_min_ratio - ) - h = int(round(math.sqrt(target_area * aspect_ratio))) - w = int(round(math.sqrt(target_area / aspect_ratio))) - if w < W and h < H: - return w, h - return None, None - - def transform_patch(self, patch): - if random.uniform(0, 1) > self.prob_flip_leftright: - patch = patch.transpose(Image.FLIP_LEFT_RIGHT) - if random.uniform(0, 1) > self.prob_rotate: - patch = patch.rotate(random.randint(-10, 10)) - return patch - - def __call__(self, img): - W, H = img.size # original image size - - # collect new patch - w, h = self.generate_wh(W, H) - if w is not None and h is not None: - x1 = random.randint(0, W - w) - y1 = random.randint(0, H - h) - new_patch = img.crop((x1, y1, x1 + w, y1 + h)) - self.patchpool.append(new_patch) - - if len(self.patchpool) < self.min_sample_size: - return img - - if random.uniform(0, 1) > self.prob_happen: - return img - - # paste a randomly selected patch on a random position - patch = random.sample(self.patchpool, 1)[0] - patchW, patchH = patch.size - x1 = random.randint(0, W - patchW) - y1 = random.randint(0, H - patchH) - patch = self.transform_patch(patch) - img.paste(patch, (x1, y1)) - - return img - - -def build_transforms( - height, - width, - transforms='random_flip', - norm_mean=[0.485, 0.456, 0.406], - norm_std=[0.229, 0.224, 0.225], - **kwargs -): - """Builds train and test transform functions. - - Args: - height (int): target image height. - width (int): target image width. - transforms (str or list of str, optional): transformations applied to model training. - Default is 'random_flip'. - norm_mean (list or None, optional): normalization mean values. Default is ImageNet means. - norm_std (list or None, optional): normalization standard deviation values. Default is - ImageNet standard deviation values. - """ - if transforms is None: - transforms = [] - - if isinstance(transforms, str): - transforms = [transforms] - - if not isinstance(transforms, list): - raise ValueError( - 'transforms must be a list of strings, but found to be {}'.format( - type(transforms) - ) - ) - - if len(transforms) > 0: - transforms = [t.lower() for t in transforms] - - if norm_mean is None or norm_std is None: - norm_mean = [0.485, 0.456, 0.406] # imagenet mean - norm_std = [0.229, 0.224, 0.225] # imagenet std - normalize = Normalize(mean=norm_mean, std=norm_std) - - print('Building train transforms ...') - transform_tr = [] - - print('+ resize to {}x{}'.format(height, width)) - transform_tr += [Resize((height, width))] - - if 'random_flip' in transforms: - print('+ random flip') - transform_tr += [RandomHorizontalFlip()] - - if 'random_crop' in transforms: - print( - '+ random crop (enlarge to {}x{} and ' - 'crop {}x{})'.format( - int(round(height * 1.125)), int(round(width * 1.125)), height, - width - ) - ) - transform_tr += [Random2DTranslation(height, width)] - - if 'random_patch' in transforms: - print('+ random patch') - transform_tr += [RandomPatch()] - - if 'color_jitter' in transforms: - print('+ color jitter') - transform_tr += [ - ColorJitter(brightness=0.2, contrast=0.15, saturation=0, hue=0) - ] - - print('+ to torch tensor of range [0, 1]') - transform_tr += [ToTensor()] - - print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) - transform_tr += [normalize] - - if 'random_erase' in transforms: - print('+ random erase') - transform_tr += [RandomErasing(mean=norm_mean)] - - transform_tr = Compose(transform_tr) - - print('Building test transforms ...') - print('+ resize to {}x{}'.format(height, width)) - print('+ to torch tensor of range [0, 1]') - print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) - - transform_te = Compose([ - Resize((height, width)), - ToTensor(), - normalize, - ]) - - return transform_tr, transform_te diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py deleted file mode 100644 index 7eca9586f7..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py +++ /dev/null @@ -1,52 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import print_function, absolute_import - -from .image import ImageSoftmaxEngine, ImageTripletEngine -from .video import VideoSoftmaxEngine, VideoTripletEngine -from .engine import Engine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py deleted file mode 100644 index 4dd250fa95..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py +++ /dev/null @@ -1,547 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import time -import numpy as np -import os.path as osp -import datetime -from collections import OrderedDict -import torch -from torch.nn import functional as F -from torch.utils.tensorboard import SummaryWriter - -from torchreid import metrics -from torchreid.utils import ( - MetricMeter, AverageMeter, re_ranking, open_all_layers, save_checkpoint, - open_specified_layers, visualize_ranked_results -) -from torchreid.losses import DeepSupervision -import os - - -class Engine(object): - r"""A generic base Engine class for both image- and video-reid. - - Args: - datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` - or ``torchreid.data.VideoDataManager``. - use_gpu (bool, optional): use gpu. Default is True. - """ - - def __init__(self, datamanager, use_gpu=False, use_npu=False): - self.datamanager = datamanager - self.train_loader = self.datamanager.train_loader - self.test_loader = self.datamanager.test_loader - self.use_gpu = use_gpu - self.use_npu = use_npu - self.writer = None - self.epoch = 0 - - self.model = None - self.optimizer = None - self.scheduler = None - - self._models = OrderedDict() - self._optims = OrderedDict() - self._scheds = OrderedDict() - - def register_model(self, name='model', model=None, optim=None, sched=None): - if self.__dict__.get('_models') is None: - raise AttributeError( - 'Cannot assign model before super().__init__() call' - ) - - if self.__dict__.get('_optims') is None: - raise AttributeError( - 'Cannot assign optim before super().__init__() call' - ) - - if self.__dict__.get('_scheds') is None: - raise AttributeError( - 'Cannot assign sched before super().__init__() call' - ) - - self._models[name] = model - self._optims[name] = optim - self._scheds[name] = sched - - def get_model_names(self, names=None): - names_real = list(self._models.keys()) - if names is not None: - if not isinstance(names, list): - names = [names] - for name in names: - assert name in names_real - return names - else: - return names_real - - def save_model(self, epoch, rank1, save_dir, is_best=False): - names = self.get_model_names() - - for name in names: - save_checkpoint( - { - 'state_dict': self._models[name].state_dict(), - 'epoch': epoch + 1, - 'rank1': rank1, - 'optimizer': self._optims[name].state_dict(), - 'scheduler': self._scheds[name].state_dict() - }, - osp.join(save_dir, name), - is_best=is_best - ) - - def set_model_mode(self, mode='train', names=None): - assert mode in ['train', 'eval', 'test'] - names = self.get_model_names(names) - - for name in names: - if mode == 'train': - self._models[name].train() - else: - self._models[name].eval() - - def get_current_lr(self, names=None): - names = self.get_model_names(names) - name = names[0] - return self._optims[name].param_groups[-1]['lr'] - - def update_lr(self, names=None): - names = self.get_model_names(names) - - for name in names: - if self._scheds[name] is not None: - self._scheds[name].step() - - def run( - self, - save_dir='log', - max_epoch=0, - start_epoch=0, - print_freq=10, - fixbase_epoch=0, - open_layers=None, - start_eval=0, - eval_freq=-1, - test_only=False, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - r"""A unified pipeline for training and evaluating a model. - - Args: - save_dir (str): directory to save model. - max_epoch (int): maximum epoch. - start_epoch (int, optional): starting epoch. Default is 0. - print_freq (int, optional): print_frequency. Default is 10. - fixbase_epoch (int, optional): number of epochs to train ``open_layers`` (new layers) - while keeping base layers frozen. Default is 0. ``fixbase_epoch`` is counted - in ``max_epoch``. - open_layers (str or list, optional): layers (attribute names) open for training. - start_eval (int, optional): from which epoch to start evaluation. Default is 0. - eval_freq (int, optional): evaluation frequency. Default is -1 (meaning evaluation - is only performed at the end of training). - test_only (bool, optional): if True, only runs evaluation on test datasets. - Default is False. - dist_metric (str, optional): distance metric used to compute distance matrix - between query and gallery. Default is "euclidean". - normalize_feature (bool, optional): performs L2 normalization on feature vectors before - computing feature distance. Default is False. - visrank (bool, optional): visualizes ranked results. Default is False. It is recommended to - enable ``visrank`` when ``test_only`` is True. The ranked images will be saved to - "save_dir/visrank_dataset", e.g. "save_dir/visrank_market1501". - visrank_topk (int, optional): top-k ranked images to be visualized. Default is 10. - use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. - Default is False. This should be enabled when using cuhk03 classic split. - ranks (list, optional): cmc ranks to be computed. Default is [1, 5, 10, 20]. - rerank (bool, optional): uses person re-ranking (by Zhong et al. CVPR'17). - Default is False. This is only enabled when test_only=True. - """ - - if visrank and not test_only: - raise ValueError( - 'visrank can be set to True only if test_only=True' - ) - - if test_only: - self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks, - rerank=rerank - ) - return - - if self.writer is None: - self.writer = SummaryWriter(log_dir=save_dir) - - time_start = time.time() - self.start_epoch = start_epoch - self.max_epoch = max_epoch - print('=> Start training') - - device_num = int(os.environ['device_num']) - device_num = 1 if device_num == -1 else device_num - batch_size = int(os.environ['batch_size']) - total_avg = 0.0 - - for self.epoch in range(self.start_epoch, self.max_epoch): - if os.environ['device_num'] != '-1' and os.environ['device_num'] != '1': - self.datamanager.train_sampler.set_epoch(self.epoch) - eve_time = self.train( - print_freq=print_freq, - fixbase_epoch=fixbase_epoch, - open_layers=open_layers - ) - total_avg += eve_time - print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / eve_time, eve_time)) - - if (self.epoch + 1) >= start_eval \ - and eval_freq > 0 \ - and (self.epoch+1) % eval_freq == 0 \ - and (self.epoch + 1) != self.max_epoch: - rank1 = self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks - ) - self.save_model(self.epoch, rank1, save_dir) - - avg_time = total_avg / (self.max_epoch - self.start_epoch) - - if self.max_epoch > 1: - print('=> Final test') - rank1 = self.test( - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks - ) - self.save_model(self.epoch, rank1, save_dir) - - elapsed = round(time.time() - time_start) - elapsed = str(datetime.timedelta(seconds=elapsed)) - print('Elapsed {}'.format(elapsed)) - - print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / avg_time, avg_time)) - - if self.writer is not None: - self.writer.close() - - def train(self, print_freq=10, fixbase_epoch=0, open_layers=None): - losses = MetricMeter() - batch_time = AverageMeter() - data_time = AverageMeter() - - self.set_model_mode('train') - - self.two_stepped_transfer_learning( - self.epoch, fixbase_epoch, open_layers - ) - - self.num_batches = len(self.train_loader) - end = time.time() - for self.batch_idx, data in enumerate(self.train_loader): - data_time.update(time.time() - end) - loss_summary = self.forward_backward(data) - batch_time.update(time.time() - end) - losses.update(loss_summary) - - if (self.batch_idx + 1) % print_freq == 0: - nb_this_epoch = self.num_batches - (self.batch_idx + 1) - nb_future_epochs = ( - self.max_epoch - (self.epoch + 1) - ) * self.num_batches - eta_seconds = batch_time.avg * (nb_this_epoch+nb_future_epochs) - eta_str = str(datetime.timedelta(seconds=int(eta_seconds))) - print( - 'epoch: [{0}/{1}][{2}/{3}]\t' - 'time {batch_time.val:.3f} ({batch_time.avg:.3f})\t' - 'data {data_time.val:.3f} ({data_time.avg:.3f})\t' - 'eta {eta}\t' - '{losses}\t' - 'lr {lr:.6f}'.format( - self.epoch + 1, - self.max_epoch, - self.batch_idx + 1, - self.num_batches, - batch_time=batch_time, - data_time=data_time, - eta=eta_str, - losses=losses, - lr=self.get_current_lr() - ) - ) - - if self.writer is not None: - n_iter = self.epoch * self.num_batches + self.batch_idx - self.writer.add_scalar('Train/time', batch_time.avg, n_iter) - self.writer.add_scalar('Train/data', data_time.avg, n_iter) - for name, meter in losses.meters.items(): - self.writer.add_scalar('Train/' + name, meter.avg, n_iter) - self.writer.add_scalar( - 'Train/lr', self.get_current_lr(), n_iter - ) - - end = time.time() - - if self.batch_idx == 4: # 前5个step不记录时间 - batch_time.reset() - - self.update_lr() - return batch_time.avg - - def forward_backward(self, data): - raise NotImplementedError - - def test( - self, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - save_dir='', - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - r"""Tests model on target datasets. - - .. note:: - - This function has been called in ``run()``. - - .. note:: - - The test pipeline implemented in this function suits both image- and - video-reid. In general, a subclass of Engine only needs to re-implement - ``extract_features()`` and ``parse_data_for_eval()`` (most of the time), - but not a must. Please refer to the source code for more details. - """ - self.set_model_mode('eval') - targets = list(self.test_loader.keys()) - - for name in targets: - domain = 'source' if name in self.datamanager.sources else 'target' - print('##### Evaluating {} ({}) #####'.format(name, domain)) - query_loader = self.test_loader[name]['query'] - gallery_loader = self.test_loader[name]['gallery'] - rank1, mAP = self._evaluate( - dataset_name=name, - query_loader=query_loader, - gallery_loader=gallery_loader, - dist_metric=dist_metric, - normalize_feature=normalize_feature, - visrank=visrank, - visrank_topk=visrank_topk, - save_dir=save_dir, - use_metric_cuhk03=use_metric_cuhk03, - ranks=ranks, - rerank=rerank - ) - - if self.writer is not None: - self.writer.add_scalar(f'Test/{name}/rank1', rank1, self.epoch) - self.writer.add_scalar(f'Test/{name}/mAP', mAP, self.epoch) - - return rank1 - - @torch.no_grad() - def _evaluate( - self, - dataset_name='', - query_loader=None, - gallery_loader=None, - dist_metric='euclidean', - normalize_feature=False, - visrank=False, - visrank_topk=10, - save_dir='', - use_metric_cuhk03=False, - ranks=[1, 5, 10, 20], - rerank=False - ): - batch_time = AverageMeter() - - def _feature_extraction(data_loader): - f_, pids_, camids_ = [], [], [] - for batch_idx, data in enumerate(data_loader): - imgs, pids, camids = self.parse_data_for_eval(data) - if self.use_gpu: - imgs = imgs.cuda() - elif self.use_npu: - imgs = imgs.npu() - end = time.time() - features = self.extract_features(imgs) - batch_time.update(time.time() - end) - features = features.cpu().clone() - f_.append(features) - pids_.extend(pids) - camids_.extend(camids) - f_ = torch.cat(f_, 0) - pids_ = np.asarray(pids_) - camids_ = np.asarray(camids_) - return f_, pids_, camids_ - - print('Extracting features from query set ...') - qf, q_pids, q_camids = _feature_extraction(query_loader) - print('Done, obtained {}-by-{} matrix'.format(qf.size(0), qf.size(1))) - - print('Extracting features from gallery set ...') - gf, g_pids, g_camids = _feature_extraction(gallery_loader) - print('Done, obtained {}-by-{} matrix'.format(gf.size(0), gf.size(1))) - - print('Speed: {:.4f} sec/batch'.format(batch_time.avg)) - - if normalize_feature: - print('Normalzing features with L2 norm ...') - qf = F.normalize(qf, p=2, dim=1) - gf = F.normalize(gf, p=2, dim=1) - - print( - 'Computing distance matrix with metric={} ...'.format(dist_metric) - ) - distmat = metrics.compute_distance_matrix(qf, gf, dist_metric) - distmat = distmat.numpy() - - if rerank: - print('Applying person re-ranking ...') - distmat_qq = metrics.compute_distance_matrix(qf, qf, dist_metric) - distmat_gg = metrics.compute_distance_matrix(gf, gf, dist_metric) - distmat = re_ranking(distmat, distmat_qq, distmat_gg) - - print('Computing CMC and mAP ...') - cmc, mAP = metrics.evaluate_rank( - distmat, - q_pids, - g_pids, - q_camids, - g_camids, - use_metric_cuhk03=use_metric_cuhk03 - ) - - print('** Results **') - print('mAP: {:.1%}'.format(mAP)) - print('CMC curve') - for r in ranks: - print('Rank-{:<3}: {:.1%}'.format(r, cmc[r - 1])) - - if visrank: - visualize_ranked_results( - distmat, - self.datamanager.fetch_test_loaders(dataset_name), - self.datamanager.data_type, - width=self.datamanager.width, - height=self.datamanager.height, - save_dir=osp.join(save_dir, 'visrank_' + dataset_name), - topk=visrank_topk - ) - - return cmc[0], mAP - - def compute_loss(self, criterion, outputs, targets): - if isinstance(outputs, (tuple, list)): - loss = DeepSupervision(criterion, outputs, targets) - else: - loss = criterion(outputs, targets) - return loss - - def extract_features(self, input): - return self.model(input) - - def parse_data_for_train(self, data): - imgs = data['img'] - pids = data['pid'] - return imgs, pids - - def parse_data_for_eval(self, data): - imgs = data['img'] - pids = data['pid'] - camids = data['camid'] - return imgs, pids, camids - - def two_stepped_transfer_learning( - self, epoch, fixbase_epoch, open_layers, model=None - ): - """Two-stepped transfer learning. - - The idea is to freeze base layers for a certain number of epochs - and then open all layers for training. - - Reference: https://arxiv.org/abs/1611.05244 - """ - model = self.model if model is None else model - if model is None: - return - - if (epoch + 1) <= fixbase_epoch and open_layers is not None: - print( - '* Only train {} (epoch: {}/{})'.format( - open_layers, epoch + 1, fixbase_epoch - ) - ) - open_specified_layers(model, open_layers) - else: - open_all_layers(model) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py deleted file mode 100644 index 0376bc80de..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py +++ /dev/null @@ -1,68 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - -from .cross_entropy_loss import CrossEntropyLoss -from .hard_mine_triplet_loss import TripletLoss - - -def DeepSupervision(criterion, xs, y): - """DeepSupervision - - Applies criterion to each element in a list. - - Args: - criterion: loss function - xs: tuple of inputs - y: ground truth - """ - loss = 0. - for x in xs: - loss += criterion(x, y) - loss /= len(xs) - return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py deleted file mode 100644 index d043691331..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py +++ /dev/null @@ -1,100 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn - - -class CrossEntropyLoss(nn.Module): - r"""Cross entropy loss with label smoothing regularizer. - - Reference: - Szegedy et al. Rethinking the Inception Architecture for Computer Vision. CVPR 2016. - - With label smoothing, the label :math:`y` for a class is computed by - - .. math:: - \begin{equation} - (1 - \eps) \times y + \frac{\eps}{K}, - \end{equation} - - where :math:`K` denotes the number of classes and :math:`\eps` is a weight. When - :math:`\eps = 0`, the loss function reduces to the normal cross entropy. - - Args: - num_classes (int): number of classes. - eps (float, optional): weight. Default is 0.1. - use_gpu (bool, optional): whether to use gpu devices. Default is True. - label_smooth (bool, optional): whether to apply label smoothing. Default is True. - """ - - def __init__(self, num_classes, eps=0.1, use_gpu=False, use_npu=False, label_smooth=True): - super(CrossEntropyLoss, self).__init__() - self.num_classes = num_classes - self.eps = eps if label_smooth else 0 - self.use_gpu = use_gpu - self.use_npu = use_npu - self.logsoftmax = nn.LogSoftmax(dim=1) - - def forward(self, inputs, targets): - """ - Args: - inputs (torch.Tensor): prediction matrix (before softmax) with - shape (batch_size, num_classes). - targets (torch.LongTensor): ground truth labels with shape (batch_size). - Each position contains the label index. - """ - log_probs = self.logsoftmax(inputs) - zeros = torch.zeros(log_probs.size()) - targets = zeros.scatter_(1, targets.unsqueeze(1).data.cpu(), 1) - if self.use_gpu: - targets = targets.cuda() - elif self.use_npu: - targets = targets.npu() - targets = (1 - self.eps) * targets + self.eps / self.num_classes - return (-targets * log_probs).mean(0).sum() diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py deleted file mode 100644 index ac2d927c92..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py +++ /dev/null @@ -1,95 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn - - -class TripletLoss(nn.Module): - """Triplet loss with hard positive/negative mining. - - Reference: - Hermans et al. In Defense of the Triplet Loss for Person Re-Identification. arXiv:1703.07737. - - Imported from ``_. - - Args: - margin (float, optional): margin for triplet. Default is 0.3. - """ - - def __init__(self, margin=0.3): - super(TripletLoss, self).__init__() - self.margin = margin - self.ranking_loss = nn.MarginRankingLoss(margin=margin) - - def forward(self, inputs, targets): - """ - Args: - inputs (torch.Tensor): feature matrix with shape (batch_size, feat_dim). - targets (torch.LongTensor): ground truth labels with shape (num_classes). - """ - n = inputs.size(0) - - # Compute pairwise distance, replace by the official when merged - dist = torch.pow(inputs, 2).sum(dim=1, keepdim=True).expand(n, n) - dist = dist + dist.t() - dist.addmm_(inputs, inputs.t(), beta=1, alpha=-2) - dist = dist.clamp(min=1e-12).sqrt() # for numerical stability - - # For each anchor, find the hardest positive and negative - mask = targets.expand(n, n).eq(targets.expand(n, n).t()) - dist_ap, dist_an = [], [] - for i in range(n): - dist_ap.append(dist[i][mask[i]].max().unsqueeze(0)) - dist_an.append(dist[i][mask[i] == 0].min().unsqueeze(0)) - dist_ap = torch.cat(dist_ap) - dist_an = torch.cat(dist_an) - - # Compute ranking hinge loss - y = torch.ones_like(dist_an) - return self.ranking_loss(dist_an, dist_ap, y) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py deleted file mode 100644 index b1c17830fa..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py +++ /dev/null @@ -1,52 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import absolute_import - -from .rank import evaluate_rank -from .accuracy import accuracy -from .distance import compute_distance_matrix diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py deleted file mode 100644 index c5145818b8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py +++ /dev/null @@ -1,84 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import - - -def accuracy(output, target, topk=(1, )): - """Computes the accuracy over the k top predictions for - the specified values of k. - - Args: - output (torch.Tensor): prediction matrix with shape (batch_size, num_classes). - target (torch.LongTensor): ground truth labels with shape (batch_size). - topk (tuple, optional): accuracy at top-k will be computed. For example, - topk=(1, 5) means accuracy at top-1 and top-5 will be computed. - - Returns: - list: accuracy at top-k. - - Examples:: - >>> from torchreid import metrics - >>> metrics.accuracy(output, target) - """ - maxk = max(topk) - batch_size = target.size(0) - - if isinstance(output, (tuple, list)): - output = output[0] - - _, pred = output.topk(maxk, 1, True, True) - pred = pred.t() - correct = pred.eq(target.view(1, -1).expand_as(pred)) - - res = [] - for k in topk: - correct_k = correct[:k].view(-1).float().sum(0, keepdim=True) - acc = correct_k.mul_(100.0 / batch_size) - res.append(acc) - - return res diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py deleted file mode 100644 index f209c03fa2..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py +++ /dev/null @@ -1,127 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import torch -from torch.nn import functional as F - - -def compute_distance_matrix(input1, input2, metric='euclidean'): - """A wrapper function for computing distance matrix. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - metric (str, optional): "euclidean" or "cosine". - Default is "euclidean". - - Returns: - torch.Tensor: distance matrix. - - Examples:: - >>> from torchreid import metrics - >>> input1 = torch.rand(10, 2048) - >>> input2 = torch.rand(100, 2048) - >>> distmat = metrics.compute_distance_matrix(input1, input2) - >>> distmat.size() # (10, 100) - """ - # check input - assert isinstance(input1, torch.Tensor) - assert isinstance(input2, torch.Tensor) - assert input1.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( - input1.dim() - ) - assert input2.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( - input2.dim() - ) - assert input1.size(1) == input2.size(1) - - if metric == 'euclidean': - distmat = euclidean_squared_distance(input1, input2) - elif metric == 'cosine': - distmat = cosine_distance(input1, input2) - else: - raise ValueError( - 'Unknown distance metric: {}. ' - 'Please choose either "euclidean" or "cosine"'.format(metric) - ) - - return distmat - - -def euclidean_squared_distance(input1, input2): - """Computes euclidean squared distance. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - - Returns: - torch.Tensor: distance matrix. - """ - m, n = input1.size(0), input2.size(0) - mat1 = torch.pow(input1, 2).sum(dim=1, keepdim=True).expand(m, n) - mat2 = torch.pow(input2, 2).sum(dim=1, keepdim=True).expand(n, m).t() - distmat = mat1 + mat2 - distmat.addmm_(input1, input2.t(), beta=1, alpha=-2) - return distmat - - -def cosine_distance(input1, input2): - """Computes cosine distance. - - Args: - input1 (torch.Tensor): 2-D feature matrix. - input2 (torch.Tensor): 2-D feature matrix. - - Returns: - torch.Tensor: distance matrix. - """ - input1_normed = F.normalize(input1, p=2, dim=1) - input2_normed = F.normalize(input2, p=2, dim=1) - distmat = 1 - torch.mm(input1_normed, input2_normed.t()) - return distmat diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py deleted file mode 100644 index 8e9fc70253..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py +++ /dev/null @@ -1,254 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, print_function, absolute_import -import numpy as np -import warnings -from collections import defaultdict - -try: - from torchreid.metrics.rank_cylib.rank_cy import evaluate_cy - IS_CYTHON_AVAI = True -except ImportError: - IS_CYTHON_AVAI = False - warnings.warn( - 'Cython evaluation (very fast so highly recommended) is ' - 'unavailable, now use python evaluation.' - ) - - -def eval_cuhk03(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): - """Evaluation with cuhk03 metric - Key: one image for each gallery identity is randomly sampled for each query identity. - Random sampling is performed num_repeats times. - """ - num_repeats = 10 - num_q, num_g = distmat.shape - - if num_g < max_rank: - max_rank = num_g - print( - 'Note: number of gallery samples is quite small, got {}'. - format(num_g) - ) - - indices = np.argsort(distmat, axis=1) - matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) - - # compute cmc curve for each query - all_cmc = [] - all_AP = [] - num_valid_q = 0. # number of valid query - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - order = indices[q_idx] - remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) - keep = np.invert(remove) - - # compute cmc curve - raw_cmc = matches[q_idx][ - keep] # binary vector, positions with value 1 are correct matches - if not np.any(raw_cmc): - # this condition is true when query identity does not appear in gallery - continue - - kept_g_pids = g_pids[order][keep] - g_pids_dict = defaultdict(list) - for idx, pid in enumerate(kept_g_pids): - g_pids_dict[pid].append(idx) - - cmc = 0. - for repeat_idx in range(num_repeats): - mask = np.zeros(len(raw_cmc), dtype=np.bool) - for _, idxs in g_pids_dict.items(): - # randomly sample one image for each gallery person - rnd_idx = np.random.choice(idxs) - mask[rnd_idx] = True - masked_raw_cmc = raw_cmc[mask] - _cmc = masked_raw_cmc.cumsum() - _cmc[_cmc > 1] = 1 - cmc += _cmc[:max_rank].astype(np.float32) - - cmc /= num_repeats - all_cmc.append(cmc) - # compute AP - num_rel = raw_cmc.sum() - tmp_cmc = raw_cmc.cumsum() - tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] - tmp_cmc = np.asarray(tmp_cmc) * raw_cmc - AP = tmp_cmc.sum() / num_rel - all_AP.append(AP) - num_valid_q += 1. - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - all_cmc = np.asarray(all_cmc).astype(np.float32) - all_cmc = all_cmc.sum(0) / num_valid_q - mAP = np.mean(all_AP) - - return all_cmc, mAP - - -def eval_market1501(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): - """Evaluation with market1501 metric - Key: for each query identity, its gallery images from the same camera view are discarded. - """ - num_q, num_g = distmat.shape - - if num_g < max_rank: - max_rank = num_g - print( - 'Note: number of gallery samples is quite small, got {}'. - format(num_g) - ) - - indices = np.argsort(distmat, axis=1) - matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) - - # compute cmc curve for each query - all_cmc = [] - all_AP = [] - num_valid_q = 0. # number of valid query - - for q_idx in range(num_q): - # get query pid and camid - q_pid = q_pids[q_idx] - q_camid = q_camids[q_idx] - - # remove gallery samples that have the same pid and camid with query - order = indices[q_idx] - remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) - keep = np.invert(remove) - - # compute cmc curve - raw_cmc = matches[q_idx][ - keep] # binary vector, positions with value 1 are correct matches - if not np.any(raw_cmc): - # this condition is true when query identity does not appear in gallery - continue - - cmc = raw_cmc.cumsum() - cmc[cmc > 1] = 1 - - all_cmc.append(cmc[:max_rank]) - num_valid_q += 1. - - # compute average precision - # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision - num_rel = raw_cmc.sum() - tmp_cmc = raw_cmc.cumsum() - tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] - tmp_cmc = np.asarray(tmp_cmc) * raw_cmc - AP = tmp_cmc.sum() / num_rel - all_AP.append(AP) - - assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' - - all_cmc = np.asarray(all_cmc).astype(np.float32) - all_cmc = all_cmc.sum(0) / num_valid_q - mAP = np.mean(all_AP) - - return all_cmc, mAP - - -def evaluate_py( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03 -): - if use_metric_cuhk03: - return eval_cuhk03( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank - ) - else: - return eval_market1501( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank - ) - - -def evaluate_rank( - distmat, - q_pids, - g_pids, - q_camids, - g_camids, - max_rank=50, - use_metric_cuhk03=False, - use_cython=True -): - """Evaluates CMC rank. - - Args: - distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). - q_pids (numpy.ndarray): 1-D array containing person identities - of each query instance. - g_pids (numpy.ndarray): 1-D array containing person identities - of each gallery instance. - q_camids (numpy.ndarray): 1-D array containing camera views under - which each query instance is captured. - g_camids (numpy.ndarray): 1-D array containing camera views under - which each gallery instance is captured. - max_rank (int, optional): maximum CMC rank to be computed. Default is 50. - use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. - Default is False. This should be enabled when using cuhk03 classic split. - use_cython (bool, optional): use cython code for evaluation. Default is True. - This is highly recommended as the cython code can speed up the cmc computation - by more than 10x. This requires Cython to be installed. - """ - if use_cython and IS_CYTHON_AVAI: - return evaluate_cy( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, - use_metric_cuhk03 - ) - else: - return evaluate_py( - distmat, q_pids, g_pids, q_camids, g_camids, max_rank, - use_metric_cuhk03 - ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py deleted file mode 100644 index 204aea72a1..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py +++ /dev/null @@ -1,425 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code source: https://github.com/pytorch/vision -""" -from __future__ import division, absolute_import -import re -from collections import OrderedDict -import torch -import torch.nn as nn -from torch.nn import functional as F -from torch.utils import model_zoo - -__all__ = [ - 'densenet121', 'densenet169', 'densenet201', 'densenet161', - 'densenet121_fc512' -] - -model_urls = { - 'densenet121': - 'https://download.pytorch.org/models/densenet121-a639ec97.pth', - 'densenet169': - 'https://download.pytorch.org/models/densenet169-b2777c0a.pth', - 'densenet201': - 'https://download.pytorch.org/models/densenet201-c1103571.pth', - 'densenet161': - 'https://download.pytorch.org/models/densenet161-8d451a50.pth', -} - - -class _DenseLayer(nn.Sequential): - - def __init__(self, num_input_features, growth_rate, bn_size, drop_rate): - super(_DenseLayer, self).__init__() - self.add_module('norm1', nn.BatchNorm2d(num_input_features)), - self.add_module('relu1', nn.ReLU(inplace=True)), - self.add_module( - 'conv1', - nn.Conv2d( - num_input_features, - bn_size * growth_rate, - kernel_size=1, - stride=1, - bias=False - ) - ), - self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate)), - self.add_module('relu2', nn.ReLU(inplace=True)), - self.add_module( - 'conv2', - nn.Conv2d( - bn_size * growth_rate, - growth_rate, - kernel_size=3, - stride=1, - padding=1, - bias=False - ) - ), - self.drop_rate = drop_rate - - def forward(self, x): - new_features = super(_DenseLayer, self).forward(x) - if self.drop_rate > 0: - new_features = F.dropout( - new_features, p=self.drop_rate, training=self.training - ) - return torch.cat([x, new_features], 1) - - -class _DenseBlock(nn.Sequential): - - def __init__( - self, num_layers, num_input_features, bn_size, growth_rate, drop_rate - ): - super(_DenseBlock, self).__init__() - for i in range(num_layers): - layer = _DenseLayer( - num_input_features + i*growth_rate, growth_rate, bn_size, - drop_rate - ) - self.add_module('denselayer%d' % (i+1), layer) - - -class _Transition(nn.Sequential): - - def __init__(self, num_input_features, num_output_features): - super(_Transition, self).__init__() - self.add_module('norm', nn.BatchNorm2d(num_input_features)) - self.add_module('relu', nn.ReLU(inplace=True)) - self.add_module( - 'conv', - nn.Conv2d( - num_input_features, - num_output_features, - kernel_size=1, - stride=1, - bias=False - ) - ) - self.add_module('pool', nn.AvgPool2d(kernel_size=2, stride=2)) - - -class DenseNet(nn.Module): - """Densely connected network. - - Reference: - Huang et al. Densely Connected Convolutional Networks. CVPR 2017. - - Public keys: - - ``densenet121``: DenseNet121. - - ``densenet169``: DenseNet169. - - ``densenet201``: DenseNet201. - - ``densenet161``: DenseNet161. - - ``densenet121_fc512``: DenseNet121 + FC. - """ - - def __init__( - self, - num_classes, - loss, - growth_rate=32, - block_config=(6, 12, 24, 16), - num_init_features=64, - bn_size=4, - drop_rate=0, - fc_dims=None, - dropout_p=None, - **kwargs - ): - - super(DenseNet, self).__init__() - self.loss = loss - - # First convolution - self.features = nn.Sequential( - OrderedDict( - [ - ( - 'conv0', - nn.Conv2d( - 3, - num_init_features, - kernel_size=7, - stride=2, - padding=3, - bias=False - ) - ), - ('norm0', nn.BatchNorm2d(num_init_features)), - ('relu0', nn.ReLU(inplace=True)), - ( - 'pool0', - nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - ), - ] - ) - ) - - # Each denseblock - num_features = num_init_features - for i, num_layers in enumerate(block_config): - block = _DenseBlock( - num_layers=num_layers, - num_input_features=num_features, - bn_size=bn_size, - growth_rate=growth_rate, - drop_rate=drop_rate - ) - self.features.add_module('denseblock%d' % (i+1), block) - num_features = num_features + num_layers*growth_rate - if i != len(block_config) - 1: - trans = _Transition( - num_input_features=num_features, - num_output_features=num_features // 2 - ) - self.features.add_module('transition%d' % (i+1), trans) - num_features = num_features // 2 - - # Final batch norm - self.features.add_module('norm5', nn.BatchNorm2d(num_features)) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.feature_dim = num_features - self.fc = self._construct_fc_layer(fc_dims, num_features, dropout_p) - - # Linear layer - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer. - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x): - f = self.features(x) - f = F.relu(f, inplace=True) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - - # '.'s are no longer allowed in module names, but pervious _DenseLayer - # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. - # They are also in the checkpoints in model_urls. This pattern is used - # to find such keys. - pattern = re.compile( - r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$' - ) - for key in list(pretrain_dict.keys()): - res = pattern.match(key) - if res: - new_key = res.group(1) + res.group(2) - pretrain_dict[new_key] = pretrain_dict[key] - del pretrain_dict[key] - - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -""" -Dense network configurations: --- -densenet121: num_init_features=64, growth_rate=32, block_config=(6, 12, 24, 16) -densenet169: num_init_features=64, growth_rate=32, block_config=(6, 12, 32, 32) -densenet201: num_init_features=64, growth_rate=32, block_config=(6, 12, 48, 32) -densenet161: num_init_features=96, growth_rate=48, block_config=(6, 12, 36, 24) -""" - - -def densenet121(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 24, 16), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet121']) - return model - - -def densenet169(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 32, 32), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet169']) - return model - - -def densenet201(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 48, 32), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet201']) - return model - - -def densenet161(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=96, - growth_rate=48, - block_config=(6, 12, 36, 24), - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet161']) - return model - - -def densenet121_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): - model = DenseNet( - num_classes=num_classes, - loss=loss, - num_init_features=64, - growth_rate=32, - block_config=(6, 12, 24, 16), - fc_dims=[512], - dropout_p=None, - **kwargs - ) - if pretrained: - init_pretrained_weights(model, model_urls['densenet121']) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py deleted file mode 100644 index 27dae2fa28..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py +++ /dev/null @@ -1,461 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -from torch import nn -from torch.nn import functional as F - -__all__ = ['HACNN'] - - -class ConvBlock(nn.Module): - """Basic convolutional block. - - convolution + batch normalization + relu. - - Args: - in_c (int): number of input channels. - out_c (int): number of output channels. - k (int or tuple): kernel size. - s (int or tuple): stride. - p (int or tuple): padding. - """ - - def __init__(self, in_c, out_c, k, s=1, p=0): - super(ConvBlock, self).__init__() - self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) - self.bn = nn.BatchNorm2d(out_c) - - def forward(self, x): - return F.relu(self.bn(self.conv(x))) - - -class InceptionA(nn.Module): - - def __init__(self, in_channels, out_channels): - super(InceptionA, self).__init__() - mid_channels = out_channels // 4 - - self.stream1 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream2 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream3 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ) - self.stream4 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1), - ConvBlock(in_channels, mid_channels, 1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - s4 = self.stream4(x) - y = torch.cat([s1, s2, s3, s4], dim=1) - return y - - -class InceptionB(nn.Module): - - def __init__(self, in_channels, out_channels): - super(InceptionB, self).__init__() - mid_channels = out_channels // 4 - - self.stream1 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), - ) - self.stream2 = nn.Sequential( - ConvBlock(in_channels, mid_channels, 1), - ConvBlock(mid_channels, mid_channels, 3, p=1), - ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), - ) - self.stream3 = nn.Sequential( - nn.MaxPool2d(3, stride=2, padding=1), - ConvBlock(in_channels, mid_channels * 2, 1), - ) - - def forward(self, x): - s1 = self.stream1(x) - s2 = self.stream2(x) - s3 = self.stream3(x) - y = torch.cat([s1, s2, s3], dim=1) - return y - - -class SpatialAttn(nn.Module): - """Spatial Attention (Sec. 3.1.I.1)""" - - def __init__(self): - super(SpatialAttn, self).__init__() - self.conv1 = ConvBlock(1, 1, 3, s=2, p=1) - self.conv2 = ConvBlock(1, 1, 1) - - def forward(self, x): - # global cross-channel averaging - x = x.mean(1, keepdim=True) - # 3-by-3 conv - x = self.conv1(x) - # bilinear resizing - x = F.upsample( - x, (x.size(2) * 2, x.size(3) * 2), - mode='bilinear', - align_corners=True - ) - # scaling conv - x = self.conv2(x) - return x - - -class ChannelAttn(nn.Module): - """Channel Attention (Sec. 3.1.I.2)""" - - def __init__(self, in_channels, reduction_rate=16): - super(ChannelAttn, self).__init__() - assert in_channels % reduction_rate == 0 - self.conv1 = ConvBlock(in_channels, in_channels // reduction_rate, 1) - self.conv2 = ConvBlock(in_channels // reduction_rate, in_channels, 1) - - def forward(self, x): - # squeeze operation (global average pooling) - x = F.avg_pool2d(x, x.size()[2:]) - # excitation operation (2 conv layers) - x = self.conv1(x) - x = self.conv2(x) - return x - - -class SoftAttn(nn.Module): - """Soft Attention (Sec. 3.1.I) - - Aim: Spatial Attention + Channel Attention - - Output: attention maps with shape identical to input. - """ - - def __init__(self, in_channels): - super(SoftAttn, self).__init__() - self.spatial_attn = SpatialAttn() - self.channel_attn = ChannelAttn(in_channels) - self.conv = ConvBlock(in_channels, in_channels, 1) - - def forward(self, x): - y_spatial = self.spatial_attn(x) - y_channel = self.channel_attn(x) - y = y_spatial * y_channel - y = torch.sigmoid(self.conv(y)) - return y - - -class HardAttn(nn.Module): - """Hard Attention (Sec. 3.1.II)""" - - def __init__(self, in_channels): - super(HardAttn, self).__init__() - self.fc = nn.Linear(in_channels, 4 * 2) - self.init_params() - - def init_params(self): - self.fc.weight.data.zero_() - self.fc.bias.data.copy_( - torch.tensor( - [0, -0.75, 0, -0.25, 0, 0.25, 0, 0.75], dtype=torch.float - ) - ) - - def forward(self, x): - # squeeze operation (global average pooling) - x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), x.size(1)) - # predict transformation parameters - theta = torch.tanh(self.fc(x)) - theta = theta.view(-1, 4, 2) - return theta - - -class HarmAttn(nn.Module): - """Harmonious Attention (Sec. 3.1)""" - - def __init__(self, in_channels): - super(HarmAttn, self).__init__() - self.soft_attn = SoftAttn(in_channels) - self.hard_attn = HardAttn(in_channels) - - def forward(self, x): - y_soft_attn = self.soft_attn(x) - theta = self.hard_attn(x) - return y_soft_attn, theta - - -class HACNN(nn.Module): - """Harmonious Attention Convolutional Neural Network. - - Reference: - Li et al. Harmonious Attention Network for Person Re-identification. CVPR 2018. - - Public keys: - - ``hacnn``: HACNN. - """ - - # Args: - # num_classes (int): number of classes to predict - # nchannels (list): number of channels AFTER concatenation - # feat_dim (int): feature dimension for a single stream - # learn_region (bool): whether to learn region features (i.e. local branch) - - def __init__( - self, - num_classes, - loss='softmax', - nchannels=[128, 256, 384], - feat_dim=512, - learn_region=True, - use_gpu=True, - **kwargs - ): - super(HACNN, self).__init__() - self.loss = loss - self.learn_region = learn_region - self.use_gpu = use_gpu - - self.conv = ConvBlock(3, 32, 3, s=2, p=1) - - # Construct Inception + HarmAttn blocks - # ============== Block 1 ============== - self.inception1 = nn.Sequential( - InceptionA(32, nchannels[0]), - InceptionB(nchannels[0], nchannels[0]), - ) - self.ha1 = HarmAttn(nchannels[0]) - - # ============== Block 2 ============== - self.inception2 = nn.Sequential( - InceptionA(nchannels[0], nchannels[1]), - InceptionB(nchannels[1], nchannels[1]), - ) - self.ha2 = HarmAttn(nchannels[1]) - - # ============== Block 3 ============== - self.inception3 = nn.Sequential( - InceptionA(nchannels[1], nchannels[2]), - InceptionB(nchannels[2], nchannels[2]), - ) - self.ha3 = HarmAttn(nchannels[2]) - - self.fc_global = nn.Sequential( - nn.Linear(nchannels[2], feat_dim), - nn.BatchNorm1d(feat_dim), - nn.ReLU(), - ) - self.classifier_global = nn.Linear(feat_dim, num_classes) - - if self.learn_region: - self.init_scale_factors() - self.local_conv1 = InceptionB(32, nchannels[0]) - self.local_conv2 = InceptionB(nchannels[0], nchannels[1]) - self.local_conv3 = InceptionB(nchannels[1], nchannels[2]) - self.fc_local = nn.Sequential( - nn.Linear(nchannels[2] * 4, feat_dim), - nn.BatchNorm1d(feat_dim), - nn.ReLU(), - ) - self.classifier_local = nn.Linear(feat_dim, num_classes) - self.feat_dim = feat_dim * 2 - else: - self.feat_dim = feat_dim - - def init_scale_factors(self): - # initialize scale factors (s_w, s_h) for four regions - self.scale_factors = [] - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - self.scale_factors.append( - torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) - ) - - def stn(self, x, theta): - """Performs spatial transform - - x: (batch, channel, height, width) - theta: (batch, 2, 3) - """ - grid = F.affine_grid(theta, x.size()) - x = F.grid_sample(x, grid) - return x - - def transform_theta(self, theta_i, region_idx): - """Transforms theta to include (s_w, s_h), resulting in (batch, 2, 3)""" - scale_factors = self.scale_factors[region_idx] - theta = torch.zeros(theta_i.size(0), 2, 3) - theta[:, :, :2] = scale_factors - theta[:, :, -1] = theta_i - if self.use_gpu: - theta = theta.cuda() - return theta - - def forward(self, x): - assert x.size(2) == 160 and x.size(3) == 64, \ - 'Input size does not match, expected (160, 64) but got ({}, {})'.format(x.size(2), x.size(3)) - x = self.conv(x) - - # ============== Block 1 ============== - # global branch - x1 = self.inception1(x) - x1_attn, x1_theta = self.ha1(x1) - x1_out = x1 * x1_attn - # local branch - if self.learn_region: - x1_local_list = [] - for region_idx in range(4): - x1_theta_i = x1_theta[:, region_idx, :] - x1_theta_i = self.transform_theta(x1_theta_i, region_idx) - x1_trans_i = self.stn(x, x1_theta_i) - x1_trans_i = F.upsample( - x1_trans_i, (24, 28), mode='bilinear', align_corners=True - ) - x1_local_i = self.local_conv1(x1_trans_i) - x1_local_list.append(x1_local_i) - - # ============== Block 2 ============== - # Block 2 - # global branch - x2 = self.inception2(x1_out) - x2_attn, x2_theta = self.ha2(x2) - x2_out = x2 * x2_attn - # local branch - if self.learn_region: - x2_local_list = [] - for region_idx in range(4): - x2_theta_i = x2_theta[:, region_idx, :] - x2_theta_i = self.transform_theta(x2_theta_i, region_idx) - x2_trans_i = self.stn(x1_out, x2_theta_i) - x2_trans_i = F.upsample( - x2_trans_i, (12, 14), mode='bilinear', align_corners=True - ) - x2_local_i = x2_trans_i + x1_local_list[region_idx] - x2_local_i = self.local_conv2(x2_local_i) - x2_local_list.append(x2_local_i) - - # ============== Block 3 ============== - # Block 3 - # global branch - x3 = self.inception3(x2_out) - x3_attn, x3_theta = self.ha3(x3) - x3_out = x3 * x3_attn - # local branch - if self.learn_region: - x3_local_list = [] - for region_idx in range(4): - x3_theta_i = x3_theta[:, region_idx, :] - x3_theta_i = self.transform_theta(x3_theta_i, region_idx) - x3_trans_i = self.stn(x2_out, x3_theta_i) - x3_trans_i = F.upsample( - x3_trans_i, (6, 7), mode='bilinear', align_corners=True - ) - x3_local_i = x3_trans_i + x2_local_list[region_idx] - x3_local_i = self.local_conv3(x3_local_i) - x3_local_list.append(x3_local_i) - - # ============== Feature generation ============== - # global branch - x_global = F.avg_pool2d(x3_out, - x3_out.size()[2:] - ).view(x3_out.size(0), x3_out.size(1)) - x_global = self.fc_global(x_global) - # local branch - if self.learn_region: - x_local_list = [] - for region_idx in range(4): - x_local_i = x3_local_list[region_idx] - x_local_i = F.avg_pool2d(x_local_i, - x_local_i.size()[2:] - ).view(x_local_i.size(0), -1) - x_local_list.append(x_local_i) - x_local = torch.cat(x_local_list, 1) - x_local = self.fc_local(x_local) - - if not self.training: - # l2 normalization before concatenation - if self.learn_region: - x_global = x_global / x_global.norm(p=2, dim=1, keepdim=True) - x_local = x_local / x_local.norm(p=2, dim=1, keepdim=True) - return torch.cat([x_global, x_local], 1) - else: - return x_global - - prelogits_global = self.classifier_global(x_global) - if self.learn_region: - prelogits_local = self.classifier_local(x_local) - - if self.loss == 'softmax': - if self.learn_region: - return (prelogits_global, prelogits_local) - else: - return prelogits_global - - elif self.loss == 'triplet': - if self.learn_region: - return (prelogits_global, prelogits_local), (x_global, x_local) - else: - return prelogits_global, x_global - - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py deleted file mode 100644 index f9d62718b3..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py +++ /dev/null @@ -1,406 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['inceptionresnetv2'] - -pretrained_settings = { - 'inceptionresnetv2': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000 - }, - 'imagenet+background': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1001 - } - } -} - - -class BasicConv2d(nn.Module): - - def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): - super(BasicConv2d, self).__init__() - self.conv = nn.Conv2d( - in_planes, - out_planes, - kernel_size=kernel_size, - stride=stride, - padding=padding, - bias=False - ) # verify bias false - self.bn = nn.BatchNorm2d( - out_planes, - eps=0.001, # value found in tensorflow - momentum=0.1, # default pytorch value - affine=True - ) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Mixed_5b(nn.Module): - - def __init__(self): - super(Mixed_5b, self).__init__() - - self.branch0 = BasicConv2d(192, 96, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(192, 48, kernel_size=1, stride=1), - BasicConv2d(48, 64, kernel_size=5, stride=1, padding=2) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(192, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), - BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(192, 64, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Block35(nn.Module): - - def __init__(self, scale=1.0): - super(Block35, self).__init__() - - self.scale = scale - - self.branch0 = BasicConv2d(320, 32, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(320, 32, kernel_size=1, stride=1), - BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(320, 32, kernel_size=1, stride=1), - BasicConv2d(32, 48, kernel_size=3, stride=1, padding=1), - BasicConv2d(48, 64, kernel_size=3, stride=1, padding=1) - ) - - self.conv2d = nn.Conv2d(128, 320, kernel_size=1, stride=1) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - out = self.conv2d(out) - out = out * self.scale + x - out = self.relu(out) - return out - - -class Mixed_6a(nn.Module): - - def __init__(self): - super(Mixed_6a, self).__init__() - - self.branch0 = BasicConv2d(320, 384, kernel_size=3, stride=2) - - self.branch1 = nn.Sequential( - BasicConv2d(320, 256, kernel_size=1, stride=1), - BasicConv2d(256, 256, kernel_size=3, stride=1, padding=1), - BasicConv2d(256, 384, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Block17(nn.Module): - - def __init__(self, scale=1.0): - super(Block17, self).__init__() - - self.scale = scale - - self.branch0 = BasicConv2d(1088, 192, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(1088, 128, kernel_size=1, stride=1), - BasicConv2d( - 128, 160, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 160, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) - ) - ) - - self.conv2d = nn.Conv2d(384, 1088, kernel_size=1, stride=1) - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - out = self.conv2d(out) - out = out * self.scale + x - out = self.relu(out) - return out - - -class Mixed_7a(nn.Module): - - def __init__(self): - super(Mixed_7a, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 384, kernel_size=3, stride=2) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 288, kernel_size=3, stride=2) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(1088, 256, kernel_size=1, stride=1), - BasicConv2d(256, 288, kernel_size=3, stride=1, padding=1), - BasicConv2d(288, 320, kernel_size=3, stride=2) - ) - - self.branch3 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Block8(nn.Module): - - def __init__(self, scale=1.0, noReLU=False): - super(Block8, self).__init__() - - self.scale = scale - self.noReLU = noReLU - - self.branch0 = BasicConv2d(2080, 192, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(2080, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 224, kernel_size=(1, 3), stride=1, padding=(0, 1) - ), - BasicConv2d( - 224, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - ) - - self.conv2d = nn.Conv2d(448, 2080, kernel_size=1, stride=1) - if not self.noReLU: - self.relu = nn.ReLU(inplace=False) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - out = self.conv2d(out) - out = out * self.scale + x - if not self.noReLU: - out = self.relu(out) - return out - - -# ---------------- -# Model Definition -# ---------------- -class InceptionResNetV2(nn.Module): - """Inception-ResNet-V2. - - Reference: - Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual - Connections on Learning. AAAI 2017. - - Public keys: - - ``inceptionresnetv2``: Inception-ResNet-V2. - """ - - def __init__(self, num_classes, loss='softmax', **kwargs): - super(InceptionResNetV2, self).__init__() - self.loss = loss - - # Modules - self.conv2d_1a = BasicConv2d(3, 32, kernel_size=3, stride=2) - self.conv2d_2a = BasicConv2d(32, 32, kernel_size=3, stride=1) - self.conv2d_2b = BasicConv2d( - 32, 64, kernel_size=3, stride=1, padding=1 - ) - self.maxpool_3a = nn.MaxPool2d(3, stride=2) - self.conv2d_3b = BasicConv2d(64, 80, kernel_size=1, stride=1) - self.conv2d_4a = BasicConv2d(80, 192, kernel_size=3, stride=1) - self.maxpool_5a = nn.MaxPool2d(3, stride=2) - self.mixed_5b = Mixed_5b() - self.repeat = nn.Sequential( - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), - Block35(scale=0.17) - ) - self.mixed_6a = Mixed_6a() - self.repeat_1 = nn.Sequential( - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), - Block17(scale=0.10), Block17(scale=0.10) - ) - self.mixed_7a = Mixed_7a() - self.repeat_2 = nn.Sequential( - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), - Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20) - ) - - self.block8 = Block8(noReLU=True) - self.conv2d_7b = BasicConv2d(2080, 1536, kernel_size=1, stride=1) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.classifier = nn.Linear(1536, num_classes) - - def load_imagenet_weights(self): - settings = pretrained_settings['inceptionresnetv2']['imagenet'] - pretrain_dict = model_zoo.load_url(settings['url']) - model_dict = self.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - self.load_state_dict(model_dict) - - def featuremaps(self, x): - x = self.conv2d_1a(x) - x = self.conv2d_2a(x) - x = self.conv2d_2b(x) - x = self.maxpool_3a(x) - x = self.conv2d_3b(x) - x = self.conv2d_4a(x) - x = self.maxpool_5a(x) - x = self.mixed_5b(x) - x = self.repeat(x) - x = self.mixed_6a(x) - x = self.repeat_1(x) - x = self.mixed_7a(x) - x = self.repeat_2(x) - x = self.block8(x) - x = self.conv2d_7b(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def inceptionresnetv2(num_classes, loss='softmax', pretrained=True, **kwargs): - model = InceptionResNetV2(num_classes=num_classes, loss=loss, **kwargs) - if pretrained: - model.load_imagenet_weights() - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py deleted file mode 100644 index 32a847a88f..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py +++ /dev/null @@ -1,428 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['inceptionv4'] -""" -Code imported from https://github.com/Cadene/pretrained-models.pytorch -""" - -pretrained_settings = { - 'inceptionv4': { - 'imagenet': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1000 - }, - 'imagenet+background': { - 'url': - 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', - 'input_space': 'RGB', - 'input_size': [3, 299, 299], - 'input_range': [0, 1], - 'mean': [0.5, 0.5, 0.5], - 'std': [0.5, 0.5, 0.5], - 'num_classes': 1001 - } - } -} - - -class BasicConv2d(nn.Module): - - def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): - super(BasicConv2d, self).__init__() - self.conv = nn.Conv2d( - in_planes, - out_planes, - kernel_size=kernel_size, - stride=stride, - padding=padding, - bias=False - ) # verify bias false - self.bn = nn.BatchNorm2d( - out_planes, - eps=0.001, # value found in tensorflow - momentum=0.1, # default pytorch value - affine=True - ) - self.relu = nn.ReLU(inplace=True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class Mixed_3a(nn.Module): - - def __init__(self): - super(Mixed_3a, self).__init__() - self.maxpool = nn.MaxPool2d(3, stride=2) - self.conv = BasicConv2d(64, 96, kernel_size=3, stride=2) - - def forward(self, x): - x0 = self.maxpool(x) - x1 = self.conv(x) - out = torch.cat((x0, x1), 1) - return out - - -class Mixed_4a(nn.Module): - - def __init__(self): - super(Mixed_4a, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(160, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(160, 64, kernel_size=1, stride=1), - BasicConv2d(64, 64, kernel_size=(1, 7), stride=1, padding=(0, 3)), - BasicConv2d(64, 64, kernel_size=(7, 1), stride=1, padding=(3, 0)), - BasicConv2d(64, 96, kernel_size=(3, 3), stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - out = torch.cat((x0, x1), 1) - return out - - -class Mixed_5a(nn.Module): - - def __init__(self): - super(Mixed_5a, self).__init__() - self.conv = BasicConv2d(192, 192, kernel_size=3, stride=2) - self.maxpool = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.conv(x) - x1 = self.maxpool(x) - out = torch.cat((x0, x1), 1) - return out - - -class Inception_A(nn.Module): - - def __init__(self): - super(Inception_A, self).__init__() - self.branch0 = BasicConv2d(384, 96, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(384, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(384, 64, kernel_size=1, stride=1), - BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), - BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(384, 96, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Reduction_A(nn.Module): - - def __init__(self): - super(Reduction_A, self).__init__() - self.branch0 = BasicConv2d(384, 384, kernel_size=3, stride=2) - - self.branch1 = nn.Sequential( - BasicConv2d(384, 192, kernel_size=1, stride=1), - BasicConv2d(192, 224, kernel_size=3, stride=1, padding=1), - BasicConv2d(224, 256, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Inception_B(nn.Module): - - def __init__(self): - super(Inception_B, self).__init__() - self.branch0 = BasicConv2d(1024, 384, kernel_size=1, stride=1) - - self.branch1 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 224, 256, kernel_size=(7, 1), stride=1, padding=(3, 0) - ) - ) - - self.branch2 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d( - 192, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), - BasicConv2d( - 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 224, 224, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), - BasicConv2d( - 224, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) - ) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(1024, 128, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - x3 = self.branch3(x) - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class Reduction_B(nn.Module): - - def __init__(self): - super(Reduction_B, self).__init__() - - self.branch0 = nn.Sequential( - BasicConv2d(1024, 192, kernel_size=1, stride=1), - BasicConv2d(192, 192, kernel_size=3, stride=2) - ) - - self.branch1 = nn.Sequential( - BasicConv2d(1024, 256, kernel_size=1, stride=1), - BasicConv2d( - 256, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) - ), - BasicConv2d( - 256, 320, kernel_size=(7, 1), stride=1, padding=(3, 0) - ), BasicConv2d(320, 320, kernel_size=3, stride=2) - ) - - self.branch2 = nn.MaxPool2d(3, stride=2) - - def forward(self, x): - x0 = self.branch0(x) - x1 = self.branch1(x) - x2 = self.branch2(x) - out = torch.cat((x0, x1, x2), 1) - return out - - -class Inception_C(nn.Module): - - def __init__(self): - super(Inception_C, self).__init__() - - self.branch0 = BasicConv2d(1536, 256, kernel_size=1, stride=1) - - self.branch1_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) - self.branch1_1a = BasicConv2d( - 384, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch1_1b = BasicConv2d( - 384, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - - self.branch2_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) - self.branch2_1 = BasicConv2d( - 384, 448, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - self.branch2_2 = BasicConv2d( - 448, 512, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch2_3a = BasicConv2d( - 512, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) - ) - self.branch2_3b = BasicConv2d( - 512, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) - ) - - self.branch3 = nn.Sequential( - nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), - BasicConv2d(1536, 256, kernel_size=1, stride=1) - ) - - def forward(self, x): - x0 = self.branch0(x) - - x1_0 = self.branch1_0(x) - x1_1a = self.branch1_1a(x1_0) - x1_1b = self.branch1_1b(x1_0) - x1 = torch.cat((x1_1a, x1_1b), 1) - - x2_0 = self.branch2_0(x) - x2_1 = self.branch2_1(x2_0) - x2_2 = self.branch2_2(x2_1) - x2_3a = self.branch2_3a(x2_2) - x2_3b = self.branch2_3b(x2_2) - x2 = torch.cat((x2_3a, x2_3b), 1) - - x3 = self.branch3(x) - - out = torch.cat((x0, x1, x2, x3), 1) - return out - - -class InceptionV4(nn.Module): - """Inception-v4. - - Reference: - Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual - Connections on Learning. AAAI 2017. - - Public keys: - - ``inceptionv4``: InceptionV4. - """ - - def __init__(self, num_classes, loss, **kwargs): - super(InceptionV4, self).__init__() - self.loss = loss - - self.features = nn.Sequential( - BasicConv2d(3, 32, kernel_size=3, stride=2), - BasicConv2d(32, 32, kernel_size=3, stride=1), - BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1), - Mixed_3a(), - Mixed_4a(), - Mixed_5a(), - Inception_A(), - Inception_A(), - Inception_A(), - Inception_A(), - Reduction_A(), # Mixed_6a - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Inception_B(), - Reduction_B(), # Mixed_7a - Inception_C(), - Inception_C(), - Inception_C() - ) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.classifier = nn.Linear(1536, num_classes) - - def forward(self, x): - f = self.features(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def inceptionv4(num_classes, loss='softmax', pretrained=True, **kwargs): - model = InceptionV4(num_classes, loss, **kwargs) - if pretrained: - model_url = pretrained_settings['inceptionv4']['imagenet']['url'] - init_pretrained_weights(model, model_url) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py deleted file mode 100644 index 0e538241f7..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py +++ /dev/null @@ -1,316 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['mlfn'] - -model_urls = { - # training epoch = 5, top1 = 51.6 - 'imagenet': - 'https://mega.nz/#!YHxAhaxC!yu9E6zWl0x5zscSouTdbZu8gdFFytDdl-RAdD2DEfpk', -} - - -class MLFNBlock(nn.Module): - - def __init__( - self, in_channels, out_channels, stride, fsm_channels, groups=32 - ): - super(MLFNBlock, self).__init__() - self.groups = groups - mid_channels = out_channels // 2 - - # Factor Modules - self.fm_conv1 = nn.Conv2d(in_channels, mid_channels, 1, bias=False) - self.fm_bn1 = nn.BatchNorm2d(mid_channels) - self.fm_conv2 = nn.Conv2d( - mid_channels, - mid_channels, - 3, - stride=stride, - padding=1, - bias=False, - groups=self.groups - ) - self.fm_bn2 = nn.BatchNorm2d(mid_channels) - self.fm_conv3 = nn.Conv2d(mid_channels, out_channels, 1, bias=False) - self.fm_bn3 = nn.BatchNorm2d(out_channels) - - # Factor Selection Module - self.fsm = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - nn.Conv2d(in_channels, fsm_channels[0], 1), - nn.BatchNorm2d(fsm_channels[0]), - nn.ReLU(inplace=True), - nn.Conv2d(fsm_channels[0], fsm_channels[1], 1), - nn.BatchNorm2d(fsm_channels[1]), - nn.ReLU(inplace=True), - nn.Conv2d(fsm_channels[1], self.groups, 1), - nn.BatchNorm2d(self.groups), - nn.Sigmoid(), - ) - - self.downsample = None - if in_channels != out_channels or stride > 1: - self.downsample = nn.Sequential( - nn.Conv2d( - in_channels, out_channels, 1, stride=stride, bias=False - ), - nn.BatchNorm2d(out_channels), - ) - - def forward(self, x): - residual = x - s = self.fsm(x) - - # reduce dimension - x = self.fm_conv1(x) - x = self.fm_bn1(x) - x = F.relu(x, inplace=True) - - # group convolution - x = self.fm_conv2(x) - x = self.fm_bn2(x) - x = F.relu(x, inplace=True) - - # factor selection - b, c = x.size(0), x.size(1) - n = c // self.groups - ss = s.repeat(1, n, 1, 1) # from (b, g, 1, 1) to (b, g*n=c, 1, 1) - ss = ss.view(b, n, self.groups, 1, 1) - ss = ss.permute(0, 2, 1, 3, 4).contiguous() - ss = ss.view(b, c, 1, 1) - x = ss * x - - # recover dimension - x = self.fm_conv3(x) - x = self.fm_bn3(x) - x = F.relu(x, inplace=True) - - if self.downsample is not None: - residual = self.downsample(residual) - - return F.relu(residual + x, inplace=True), s - - -class MLFN(nn.Module): - """Multi-Level Factorisation Net. - - Reference: - Chang et al. Multi-Level Factorisation Net for - Person Re-Identification. CVPR 2018. - - Public keys: - - ``mlfn``: MLFN (Multi-Level Factorisation Net). - """ - - def __init__( - self, - num_classes, - loss='softmax', - groups=32, - channels=[64, 256, 512, 1024, 2048], - embed_dim=1024, - **kwargs - ): - super(MLFN, self).__init__() - self.loss = loss - self.groups = groups - - # first convolutional layer - self.conv1 = nn.Conv2d(3, channels[0], 7, stride=2, padding=3) - self.bn1 = nn.BatchNorm2d(channels[0]) - self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) - - # main body - self.feature = nn.ModuleList( - [ - # layer 1-3 - MLFNBlock(channels[0], channels[1], 1, [128, 64], self.groups), - MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), - MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), - # layer 4-7 - MLFNBlock( - channels[1], channels[2], 2, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - MLFNBlock( - channels[2], channels[2], 1, [256, 128], self.groups - ), - # layer 8-13 - MLFNBlock( - channels[2], channels[3], 2, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[3], channels[3], 1, [512, 128], self.groups - ), - # layer 14-16 - MLFNBlock( - channels[3], channels[4], 2, [512, 128], self.groups - ), - MLFNBlock( - channels[4], channels[4], 1, [512, 128], self.groups - ), - MLFNBlock( - channels[4], channels[4], 1, [512, 128], self.groups - ), - ] - ) - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - - # projection functions - self.fc_x = nn.Sequential( - nn.Conv2d(channels[4], embed_dim, 1, bias=False), - nn.BatchNorm2d(embed_dim), - nn.ReLU(inplace=True), - ) - self.fc_s = nn.Sequential( - nn.Conv2d(self.groups * 16, embed_dim, 1, bias=False), - nn.BatchNorm2d(embed_dim), - nn.ReLU(inplace=True), - ) - - self.classifier = nn.Linear(embed_dim, num_classes) - - self.init_params() - - def init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = F.relu(x, inplace=True) - x = self.maxpool(x) - - s_hat = [] - for block in self.feature: - x, s = block(x) - s_hat.append(s) - s_hat = torch.cat(s_hat, 1) - - x = self.global_avgpool(x) - x = self.fc_x(x) - s_hat = self.fc_s(s_hat) - - v = (x+s_hat) * 0.5 - v = v.view(v.size(0), -1) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError('Unsupported loss: {}'.format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def mlfn(num_classes, loss='softmax', pretrained=True, **kwargs): - model = MLFN(num_classes, loss, **kwargs) - if pretrained: - # init_pretrained_weights(model, model_urls['imagenet']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['imagenet']) - ) - return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py deleted file mode 100644 index 690dade1bb..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py +++ /dev/null @@ -1,321 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from __future__ import division, absolute_import -import torch.utils.model_zoo as model_zoo -from torch import nn -from torch.nn import functional as F - -__all__ = ['mobilenetv2_x1_0', 'mobilenetv2_x1_4'] - -model_urls = { - # 1.0: top-1 71.3 - 'mobilenetv2_x1_0': - 'https://mega.nz/#!NKp2wAIA!1NH1pbNzY_M2hVk_hdsxNM1NUOWvvGPHhaNr-fASF6c', - # 1.4: top-1 73.9 - 'mobilenetv2_x1_4': - 'https://mega.nz/#!RGhgEIwS!xN2s2ZdyqI6vQ3EwgmRXLEW3khr9tpXg96G9SUJugGk', -} - - -class ConvBlock(nn.Module): - """Basic convolutional block. - - convolution (bias discarded) + batch normalization + relu6. - - Args: - in_c (int): number of input channels. - out_c (int): number of output channels. - k (int or tuple): kernel size. - s (int or tuple): stride. - p (int or tuple): padding. - g (int): number of blocked connections from input channels - to output channels (default: 1). - """ - - def __init__(self, in_c, out_c, k, s=1, p=0, g=1): - super(ConvBlock, self).__init__() - self.conv = nn.Conv2d( - in_c, out_c, k, stride=s, padding=p, bias=False, groups=g - ) - self.bn = nn.BatchNorm2d(out_c) - - def forward(self, x): - return F.relu6(self.bn(self.conv(x))) - - -class Bottleneck(nn.Module): - - def __init__(self, in_channels, out_channels, expansion_factor, stride=1): - super(Bottleneck, self).__init__() - mid_channels = in_channels * expansion_factor - self.use_residual = stride == 1 and in_channels == out_channels - self.conv1 = ConvBlock(in_channels, mid_channels, 1) - self.dwconv2 = ConvBlock( - mid_channels, mid_channels, 3, stride, 1, g=mid_channels - ) - self.conv3 = nn.Sequential( - nn.Conv2d(mid_channels, out_channels, 1, bias=False), - nn.BatchNorm2d(out_channels), - ) - - def forward(self, x): - m = self.conv1(x) - m = self.dwconv2(m) - m = self.conv3(m) - if self.use_residual: - return x + m - else: - return m - - -class MobileNetV2(nn.Module): - """MobileNetV2. - - Reference: - Sandler et al. MobileNetV2: Inverted Residuals and - Linear Bottlenecks. CVPR 2018. - - Public keys: - - ``mobilenetv2_x1_0``: MobileNetV2 x1.0. - - ``mobilenetv2_x1_4``: MobileNetV2 x1.4. - """ - - def __init__( - self, - num_classes, - width_mult=1, - loss='softmax', - fc_dims=None, - dropout_p=None, - **kwargs - ): - super(MobileNetV2, self).__init__() - self.loss = loss - self.in_channels = int(32 * width_mult) - self.feature_dim = int(1280 * width_mult) if width_mult > 1 else 1280 - - # construct layers - self.conv1 = ConvBlock(3, self.in_channels, 3, s=2, p=1) - self.conv2 = self._make_layer( - Bottleneck, 1, int(16 * width_mult), 1, 1 - ) - self.conv3 = self._make_layer( - Bottleneck, 6, int(24 * width_mult), 2, 2 - ) - self.conv4 = self._make_layer( - Bottleneck, 6, int(32 * width_mult), 3, 2 - ) - self.conv5 = self._make_layer( - Bottleneck, 6, int(64 * width_mult), 4, 2 - ) - self.conv6 = self._make_layer( - Bottleneck, 6, int(96 * width_mult), 3, 1 - ) - self.conv7 = self._make_layer( - Bottleneck, 6, int(160 * width_mult), 3, 2 - ) - self.conv8 = self._make_layer( - Bottleneck, 6, int(320 * width_mult), 1, 1 - ) - self.conv9 = ConvBlock(self.in_channels, self.feature_dim, 1) - - self.global_avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = self._construct_fc_layer( - fc_dims, self.feature_dim, dropout_p - ) - self.classifier = nn.Linear(self.feature_dim, num_classes) - - self._init_params() - - def _make_layer(self, block, t, c, n, s): - # t: expansion factor - # c: output channels - # n: number of blocks - # s: stride for first layer - layers = [] - layers.append(block(self.in_channels, c, t, s)) - self.in_channels = c - for i in range(1, n): - layers.append(block(self.in_channels, c, t)) - return nn.Sequential(*layers) - - def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): - """Constructs fully connected layer. - - Args: - fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed - input_dim (int): input dimension - dropout_p (float): dropout probability, if None, dropout is unused - """ - if fc_dims is None: - self.feature_dim = input_dim - return None - - assert isinstance( - fc_dims, (list, tuple) - ), 'fc_dims must be either list or tuple, but got {}'.format( - type(fc_dims) - ) - - layers = [] - for dim in fc_dims: - layers.append(nn.Linear(input_dim, dim)) - layers.append(nn.BatchNorm1d(dim)) - layers.append(nn.ReLU(inplace=True)) - if dropout_p is not None: - layers.append(nn.Dropout(p=dropout_p)) - input_dim = dim - - self.feature_dim = fc_dims[-1] - - return nn.Sequential(*layers) - - def _init_params(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_( - m.weight, mode='fan_out', nonlinearity='relu' - ) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm1d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def featuremaps(self, x): - x = self.conv1(x) - x = self.conv2(x) - x = self.conv3(x) - x = self.conv4(x) - x = self.conv5(x) - x = self.conv6(x) - x = self.conv7(x) - x = self.conv8(x) - x = self.conv9(x) - return x - - def forward(self, x): - f = self.featuremaps(x) - v = self.global_avgpool(f) - v = v.view(v.size(0), -1) - - if self.fc is not None: - v = self.fc(v) - - if not self.training: - return v - - y = self.classifier(v) - - if self.loss == 'softmax': - return y - elif self.loss == 'triplet': - return y, v - else: - raise KeyError("Unsupported loss: {}".format(self.loss)) - - -def init_pretrained_weights(model, model_url): - """Initializes model with pretrained weights. - - Layers that don't match with pretrained layers in name or size are kept unchanged. - """ - pretrain_dict = model_zoo.load_url(model_url) - model_dict = model.state_dict() - pretrain_dict = { - k: v - for k, v in pretrain_dict.items() - if k in model_dict and model_dict[k].size() == v.size() - } - model_dict.update(pretrain_dict) - model.load_state_dict(model_dict) - - -def mobilenetv2_x1_0(num_classes, loss, pretrained=True, **kwargs): - model = MobileNetV2( - num_classes, - loss=loss, - width_mult=1, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - # init_pretrained_weights(model, model_urls['mobilenetv2_x1_0']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['mobilenetv2_x1_0']) - ) - return model - - -def mobilenetv2_x1_4(num_classes, loss, pretrained=True, **kwargs): - model = MobileNetV2( - num_classes, - loss=loss, - width_mult=1.4, - fc_dims=None, - dropout_p=None, - **kwargs - ) - if pretrained: - # init_pretrained_weights(model, model_urls['mobilenetv2_x1_4']) - import warnings - warnings.warn( - 'The imagenet pretrained weights need to be manually downloaded from {}' - .format(model_urls['mobilenetv2_x1_4']) - ) - return model -- Gitee From aa8d22405e4365e590d43f3b65e570a2386f3845 Mon Sep 17 00:00:00 2001 From: BYT <1044237391@qq.com> Date: Thu, 9 Jun 2022 18:17:23 +0800 Subject: [PATCH 13/31] id --- .../OSNet/torchreid/__init__.py | 56 + .../OSNet/torchreid/data/__init__.py | 54 + .../OSNet/torchreid/data/datamanager.py | 609 +++++++++ .../OSNet/torchreid/data/datasets/__init__.py | 166 +++ .../OSNet/torchreid/data/datasets/dataset.py | 529 ++++++++ .../torchreid/data/datasets/image/__init__.py | 62 + .../torchreid/data/datasets/image/cuhk01.py | 184 +++ .../torchreid/data/datasets/image/cuhk02.py | 144 ++ .../torchreid/data/datasets/image/cuhk03.py | 354 +++++ .../torchreid/data/datasets/image/cuhksysu.py | 107 ++ .../data/datasets/image/dukemtmcreid.py | 115 ++ .../torchreid/data/datasets/image/grid.py | 178 +++ .../torchreid/data/datasets/image/ilids.py | 182 +++ .../data/datasets/image/market1501.py | 133 ++ .../torchreid/data/datasets/image/msmt17.py | 145 ++ .../torchreid/data/datasets/image/prid.py | 154 +++ .../data/datasets/image/sensereid.py | 117 ++ .../data/datasets/image/university1652.py | 157 +++ .../torchreid/data/datasets/image/viper.py | 175 +++ .../torchreid/data/datasets/video/__init__.py | 53 + .../data/datasets/video/dukemtmcvidreid.py | 175 +++ .../torchreid/data/datasets/video/ilidsvid.py | 190 +++ .../torchreid/data/datasets/video/mars.py | 180 +++ .../torchreid/data/datasets/video/prid2011.py | 127 ++ .../OSNet/torchreid/data/sampler.py | 292 ++++ .../OSNet/torchreid/data/transforms.py | 373 ++++++ .../OSNet/torchreid/engine/__init__.py | 52 + .../OSNet/torchreid/engine/engine.py | 547 ++++++++ .../OSNet/torchreid/engine/image/__init__.py | 51 + .../OSNet/torchreid/engine/image/softmax.py | 157 +++ .../OSNet/torchreid/engine/image/triplet.py | 169 +++ .../OSNet/torchreid/engine/video/__init__.py | 51 + .../OSNet/torchreid/engine/video/softmax.py | 156 +++ .../OSNet/torchreid/engine/video/triplet.py | 169 +++ .../OSNet/torchreid/losses/__init__.py | 68 + .../torchreid/losses/cross_entropy_loss.py | 100 ++ .../losses/hard_mine_triplet_loss.py | 95 ++ .../OSNet/torchreid/metrics/__init__.py | 52 + .../OSNet/torchreid/metrics/accuracy.py | 84 ++ .../OSNet/torchreid/metrics/distance.py | 127 ++ .../OSNet/torchreid/metrics/rank.py | 254 ++++ .../torchreid/metrics/rank_cylib/Makefile | 6 + .../torchreid/metrics/rank_cylib/__init__.py | 47 + .../torchreid/metrics/rank_cylib/rank_cy.pyx | 251 ++++ .../torchreid/metrics/rank_cylib/setup.py | 73 + .../metrics/rank_cylib/test_cython.py | 130 ++ .../OSNet/torchreid/models/__init__.py | 166 +++ .../OSNet/torchreid/models/densenet.py | 425 ++++++ .../OSNet/torchreid/models/hacnn.py | 461 +++++++ .../torchreid/models/inceptionresnetv2.py | 406 ++++++ .../OSNet/torchreid/models/inceptionv4.py | 428 ++++++ .../OSNet/torchreid/models/mlfn.py | 316 +++++ .../OSNet/torchreid/models/mobilenetv2.py | 321 +++++ .../OSNet/torchreid/models/mudeep.py | 253 ++++ .../OSNet/torchreid/models/nasnet.py | 1178 +++++++++++++++++ .../OSNet/torchreid/models/osnet.py | 645 +++++++++ .../OSNet/torchreid/models/osnet_ain.py | 588 ++++++++ .../OSNet/torchreid/models/pcb.py | 361 +++++ .../OSNet/torchreid/models/resnet.py | 576 ++++++++ .../OSNet/torchreid/models/resnet_ibn_a.py | 334 +++++ .../OSNet/torchreid/models/resnet_ibn_b.py | 319 +++++ .../OSNet/torchreid/models/resnetmid.py | 354 +++++ .../OSNet/torchreid/models/senet.py | 735 ++++++++++ .../OSNet/torchreid/models/shufflenet.py | 245 ++++ .../OSNet/torchreid/models/shufflenetv2.py | 307 +++++ .../OSNet/torchreid/models/squeezenet.py | 281 ++++ .../OSNet/torchreid/models/xception.py | 391 ++++++ .../OSNet/torchreid/optim/__init__.py | 51 + .../OSNet/torchreid/optim/lr_scheduler.py | 115 ++ .../OSNet/torchreid/optim/optimizer.py | 218 +++ .../OSNet/torchreid/optim/radam.py | 376 ++++++ .../OSNet/torchreid/utils/__init__.py | 57 + .../OSNet/torchreid/utils/avgmeter.py | 120 ++ .../torchreid/utils/feature_extractor.py | 199 +++ .../OSNet/torchreid/utils/loggers.py | 193 +++ .../OSNet/torchreid/utils/model_complexity.py | 410 ++++++ .../OSNet/torchreid/utils/reidtools.py | 201 +++ .../OSNet/torchreid/utils/rerank.py | 157 +++ .../OSNet/torchreid/utils/tools.py | 189 +++ .../OSNet/torchreid/utils/torchtools.py | 363 +++++ 80 files changed, 19389 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py new file mode 100644 index 0000000000..40669b944c --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/__init__.py @@ -0,0 +1,56 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from torchreid import data, optim, utils, engine, losses, models, metrics + +__version__ = '1.4.0' +__author__ = 'Kaiyang Zhou' +__homepage__ = 'https://kaiyangzhou.github.io/' +__description__ = 'Deep learning person re-identification in PyTorch' +__url__ = 'https://github.com/KaiyangZhou/deep-person-reid' diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py new file mode 100644 index 0000000000..577ed45750 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/__init__.py @@ -0,0 +1,54 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .datasets import ( + Dataset, ImageDataset, VideoDataset, register_image_dataset, + register_video_dataset +) +from .datamanager import ImageDataManager, VideoDataManager diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py new file mode 100644 index 0000000000..009d03802f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datamanager.py @@ -0,0 +1,609 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch + +from torchreid.data.sampler import build_train_sampler +from torchreid.data.datasets import init_image_dataset, init_video_dataset +from torchreid.data.transforms import build_transforms + + +class DataManager(object): + r"""Base data manager. + + Args: + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + """ + + def __init__( + self, + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + norm_mean=None, + norm_std=None, + use_gpu=False + ): + self.sources = sources + self.targets = targets + self.height = height + self.width = width + + if self.sources is None: + raise ValueError('sources must not be None') + + if isinstance(self.sources, str): + self.sources = [self.sources] + + if self.targets is None: + self.targets = self.sources + + if isinstance(self.targets, str): + self.targets = [self.targets] + + self.transform_tr, self.transform_te = build_transforms( + self.height, + self.width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std + ) + + self.use_gpu = (torch.cuda.is_available() and use_gpu) + + @property + def num_train_pids(self): + """Returns the number of training person identities.""" + return self._num_train_pids + + @property + def num_train_cams(self): + """Returns the number of training cameras.""" + return self._num_train_cams + + def fetch_test_loaders(self, name): + """Returns query and gallery of a test dataset, each containing + tuples of (img_path(s), pid, camid). + + Args: + name (str): dataset name. + """ + query_loader = self.test_dataset[name]['query'] + gallery_loader = self.test_dataset[name]['gallery'] + return query_loader, gallery_loader + + def preprocess_pil_img(self, img): + """Transforms a PIL image to torch tensor for testing.""" + return self.transform_te(img) + + +class ImageDataManager(DataManager): + r"""Image data manager. + + Args: + root (str): root path to datasets. + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + k_tfm (int): number of times to apply augmentation to an image + independently. If k_tfm > 1, the transform function will be + applied k_tfm times to an image. This variable will only be + useful for training and is currently valid for image datasets only. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + split_id (int, optional): split id (*0-based*). Default is 0. + combineall (bool, optional): combine train, query and gallery in a dataset for + training. Default is False. + load_train_targets (bool, optional): construct train-loader for target datasets. + Default is False. This is useful for domain adaptation research. + batch_size_train (int, optional): number of images in a training batch. Default is 32. + batch_size_test (int, optional): number of images in a test batch. Default is 32. + workers (int, optional): number of workers. Default is 4. + num_instances (int, optional): number of instances per identity in a batch. + Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + train_sampler (str, optional): sampler. Default is RandomSampler. + train_sampler_t (str, optional): sampler for target train loader. Default is RandomSampler. + cuhk03_labeled (bool, optional): use cuhk03 labeled images. + Default is False (defaul is to use detected images). + cuhk03_classic_split (bool, optional): use the classic split in cuhk03. + Default is False. + market1501_500k (bool, optional): add 500K distractors to the gallery + set in market1501. Default is False. + + Examples:: + + datamanager = torchreid.data.ImageDataManager( + root='path/to/reid-data', + sources='market1501', + height=256, + width=128, + batch_size_train=32, + batch_size_test=100 + ) + + # return train loader of source data + train_loader = datamanager.train_loader + + # return test loader of target data + test_loader = datamanager.test_loader + + # return train loader of target data + train_loader_t = datamanager.train_loader_t + """ + data_type = 'image' + + def __init__( + self, + root='', + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + k_tfm=1, + norm_mean=None, + norm_std=None, + use_gpu=True, + split_id=0, + combineall=False, + load_train_targets=False, + batch_size_train=32, + batch_size_test=32, + workers=4, + num_instances=4, + num_cams=1, + num_datasets=1, + train_sampler='RandomSampler', + train_sampler_t='RandomSampler', + cuhk03_labeled=False, + cuhk03_classic_split=False, + market1501_500k=False, + device_num=-1 + ): + + super(ImageDataManager, self).__init__( + sources=sources, + targets=targets, + height=height, + width=width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std, + use_gpu=use_gpu + ) + + print('=> Loading train (source) dataset') + trainset = [] + for name in self.sources: + trainset_ = init_image_dataset( + name, + transform=self.transform_tr, + k_tfm=k_tfm, + mode='train', + combineall=combineall, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + trainset.append(trainset_) + trainset = sum(trainset) + + self._num_train_pids = trainset.num_train_pids + self._num_train_cams = trainset.num_train_cams + + if device_num == -1 or device_num == 1: + self.train_sampler = build_train_sampler( + trainset.train, + train_sampler, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ) + + else: + self.train_sampler = torch.utils.data.distributed.DistributedSampler(trainset.train) + + self.train_loader = torch.utils.data.DataLoader( + trainset, + sampler=self.train_sampler, + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + # pin_memory=self.use_gpu, + pin_memory=True, + drop_last=True + ) + + self.train_loader_t = None + if load_train_targets: + # check if sources and targets are identical + assert len(set(self.sources) & set(self.targets)) == 0, \ + 'sources={} and targets={} must not have overlap'.format(self.sources, self.targets) + + print('=> Loading train (target) dataset') + trainset_t = [] + for name in self.targets: + trainset_t_ = init_image_dataset( + name, + transform=self.transform_tr, + k_tfm=k_tfm, + mode='train', + combineall=False, # only use the training data + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + trainset_t.append(trainset_t_) + trainset_t = sum(trainset_t) + + self.train_loader_t = torch.utils.data.DataLoader( + trainset_t, + sampler=build_train_sampler( + trainset_t.train, + train_sampler_t, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ), + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=True + ) + + print('=> Loading test (target) dataset') + self.test_loader = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + self.test_dataset = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + + for name in self.targets: + # build query loader + queryset = init_image_dataset( + name, + transform=self.transform_te, + mode='query', + combineall=combineall, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + self.test_loader[name]['query'] = torch.utils.data.DataLoader( + queryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=True, + drop_last=False + ) + + # build gallery loader + galleryset = init_image_dataset( + name, + transform=self.transform_te, + mode='gallery', + combineall=combineall, + verbose=False, + root=root, + split_id=split_id, + cuhk03_labeled=cuhk03_labeled, + cuhk03_classic_split=cuhk03_classic_split, + market1501_500k=market1501_500k + ) + self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( + galleryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=True, + drop_last=False + ) + + self.test_dataset[name]['query'] = queryset.query + self.test_dataset[name]['gallery'] = galleryset.gallery + + print('\n') + print(' **************** Summary ****************') + print(' source : {}'.format(self.sources)) + print(' # source datasets : {}'.format(len(self.sources))) + print(' # source ids : {}'.format(self.num_train_pids)) + print(' # source images : {}'.format(len(trainset))) + print(' # source cameras : {}'.format(self.num_train_cams)) + if load_train_targets: + print( + ' # target images : {} (unlabeled)'.format(len(trainset_t)) + ) + print(' target : {}'.format(self.targets)) + print(' *****************************************') + print('\n') + + +class VideoDataManager(DataManager): + r"""Video data manager. + + Args: + root (str): root path to datasets. + sources (str or list): source dataset(s). + targets (str or list, optional): target dataset(s). If not given, + it equals to ``sources``. + height (int, optional): target image height. Default is 256. + width (int, optional): target image width. Default is 128. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): data mean. Default is None (use imagenet mean). + norm_std (list or None, optional): data std. Default is None (use imagenet std). + use_gpu (bool, optional): use gpu. Default is True. + split_id (int, optional): split id (*0-based*). Default is 0. + combineall (bool, optional): combine train, query and gallery in a dataset for + training. Default is False. + batch_size_train (int, optional): number of tracklets in a training batch. Default is 3. + batch_size_test (int, optional): number of tracklets in a test batch. Default is 3. + workers (int, optional): number of workers. Default is 4. + num_instances (int, optional): number of instances per identity in a batch. + Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + train_sampler (str, optional): sampler. Default is RandomSampler. + seq_len (int, optional): how many images to sample in a tracklet. Default is 15. + sample_method (str, optional): how to sample images in a tracklet. Default is "evenly". + Choices are ["evenly", "random", "all"]. "evenly" and "random" will sample ``seq_len`` + images in a tracklet while "all" samples all images in a tracklet, where the batch size + needs to be set to 1. + + Examples:: + + datamanager = torchreid.data.VideoDataManager( + root='path/to/reid-data', + sources='mars', + height=256, + width=128, + batch_size_train=3, + batch_size_test=3, + seq_len=15, + sample_method='evenly' + ) + + # return train loader of source data + train_loader = datamanager.train_loader + + # return test loader of target data + test_loader = datamanager.test_loader + + .. note:: + The current implementation only supports image-like training. Therefore, each image in a + sampled tracklet will undergo independent transformation functions. To achieve tracklet-aware + training, you need to modify the transformation functions for video reid such that each function + applies the same operation to all images in a tracklet to keep consistency. + """ + data_type = 'video' + + def __init__( + self, + root='', + sources=None, + targets=None, + height=256, + width=128, + transforms='random_flip', + norm_mean=None, + norm_std=None, + use_gpu=True, + split_id=0, + combineall=False, + batch_size_train=3, + batch_size_test=3, + workers=4, + num_instances=4, + num_cams=1, + num_datasets=1, + train_sampler='RandomSampler', + seq_len=15, + sample_method='evenly' + ): + + super(VideoDataManager, self).__init__( + sources=sources, + targets=targets, + height=height, + width=width, + transforms=transforms, + norm_mean=norm_mean, + norm_std=norm_std, + use_gpu=use_gpu + ) + + print('=> Loading train (source) dataset') + trainset = [] + for name in self.sources: + trainset_ = init_video_dataset( + name, + transform=self.transform_tr, + mode='train', + combineall=combineall, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + trainset.append(trainset_) + trainset = sum(trainset) + + self._num_train_pids = trainset.num_train_pids + self._num_train_cams = trainset.num_train_cams + + train_sampler = build_train_sampler( + trainset.train, + train_sampler, + batch_size=batch_size_train, + num_instances=num_instances, + num_cams=num_cams, + num_datasets=num_datasets + ) + + self.train_loader = torch.utils.data.DataLoader( + trainset, + sampler=train_sampler, + batch_size=batch_size_train, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=True + ) + + print('=> Loading test (target) dataset') + self.test_loader = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + self.test_dataset = { + name: { + 'query': None, + 'gallery': None + } + for name in self.targets + } + + for name in self.targets: + # build query loader + queryset = init_video_dataset( + name, + transform=self.transform_te, + mode='query', + combineall=combineall, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + self.test_loader[name]['query'] = torch.utils.data.DataLoader( + queryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=False + ) + + # build gallery loader + galleryset = init_video_dataset( + name, + transform=self.transform_te, + mode='gallery', + combineall=combineall, + verbose=False, + root=root, + split_id=split_id, + seq_len=seq_len, + sample_method=sample_method + ) + self.test_loader[name]['gallery'] = torch.utils.data.DataLoader( + galleryset, + batch_size=batch_size_test, + shuffle=False, + num_workers=workers, + pin_memory=self.use_gpu, + drop_last=False + ) + + self.test_dataset[name]['query'] = queryset.query + self.test_dataset[name]['gallery'] = galleryset.gallery + + print('\n') + print(' **************** Summary ****************') + print(' source : {}'.format(self.sources)) + print(' # source datasets : {}'.format(len(self.sources))) + print(' # source ids : {}'.format(self.num_train_pids)) + print(' # source tracklets : {}'.format(len(trainset))) + print(' # source cameras : {}'.format(self.num_train_cams)) + print(' target : {}'.format(self.targets)) + print(' *****************************************') + print('\n') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py new file mode 100644 index 0000000000..2b18ca7207 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/__init__.py @@ -0,0 +1,166 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .image import ( + GRID, PRID, CUHK01, CUHK02, CUHK03, MSMT17, CUHKSYSU, VIPeR, SenseReID, + Market1501, DukeMTMCreID, University1652, iLIDS +) +from .video import PRID2011, Mars, DukeMTMCVidReID, iLIDSVID +from .dataset import Dataset, ImageDataset, VideoDataset + +__image_datasets = { + 'market1501': Market1501, + 'cuhk03': CUHK03, + 'dukemtmcreid': DukeMTMCreID, + 'msmt17': MSMT17, + 'viper': VIPeR, + 'grid': GRID, + 'cuhk01': CUHK01, + 'ilids': iLIDS, + 'sensereid': SenseReID, + 'prid': PRID, + 'cuhk02': CUHK02, + 'university1652': University1652, + 'cuhksysu': CUHKSYSU +} + +__video_datasets = { + 'mars': Mars, + 'ilidsvid': iLIDSVID, + 'prid2011': PRID2011, + 'dukemtmcvidreid': DukeMTMCVidReID +} + + +def init_image_dataset(name, **kwargs): + """Initializes an image dataset.""" + avai_datasets = list(__image_datasets.keys()) + if name not in avai_datasets: + raise ValueError( + 'Invalid dataset name. Received "{}", ' + 'but expected to be one of {}'.format(name, avai_datasets) + ) + return __image_datasets[name](**kwargs) + + +def init_video_dataset(name, **kwargs): + """Initializes a video dataset.""" + avai_datasets = list(__video_datasets.keys()) + if name not in avai_datasets: + raise ValueError( + 'Invalid dataset name. Received "{}", ' + 'but expected to be one of {}'.format(name, avai_datasets) + ) + return __video_datasets[name](**kwargs) + + +def register_image_dataset(name, dataset): + """Registers a new image dataset. + + Args: + name (str): key corresponding to the new dataset. + dataset (Dataset): the new dataset class. + + Examples:: + + import torchreid + import NewDataset + torchreid.data.register_image_dataset('new_dataset', NewDataset) + # single dataset case + datamanager = torchreid.data.ImageDataManager( + root='reid-data', + sources='new_dataset' + ) + # multiple dataset case + datamanager = torchreid.data.ImageDataManager( + root='reid-data', + sources=['new_dataset', 'dukemtmcreid'] + ) + """ + global __image_datasets + curr_datasets = list(__image_datasets.keys()) + if name in curr_datasets: + raise ValueError( + 'The given name already exists, please choose ' + 'another name excluding {}'.format(curr_datasets) + ) + __image_datasets[name] = dataset + + +def register_video_dataset(name, dataset): + """Registers a new video dataset. + + Args: + name (str): key corresponding to the new dataset. + dataset (Dataset): the new dataset class. + + Examples:: + + import torchreid + import NewDataset + torchreid.data.register_video_dataset('new_dataset', NewDataset) + # single dataset case + datamanager = torchreid.data.VideoDataManager( + root='reid-data', + sources='new_dataset' + ) + # multiple dataset case + datamanager = torchreid.data.VideoDataManager( + root='reid-data', + sources=['new_dataset', 'ilidsvid'] + ) + """ + global __video_datasets + curr_datasets = list(__video_datasets.keys()) + if name in curr_datasets: + raise ValueError( + 'The given name already exists, please choose ' + 'another name excluding {}'.format(curr_datasets) + ) + __video_datasets[name] = dataset diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py new file mode 100644 index 0000000000..f85cb3533e --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/dataset.py @@ -0,0 +1,529 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import copy +import numpy as np +import os.path as osp +import tarfile +import zipfile +import torch + +from torchreid.utils import read_image, download_url, mkdir_if_missing + + +class Dataset(object): + """An abstract class representing a Dataset. + + This is the base class for ``ImageDataset`` and ``VideoDataset``. + + Args: + train (list): contains tuples of (img_path(s), pid, camid). + query (list): contains tuples of (img_path(s), pid, camid). + gallery (list): contains tuples of (img_path(s), pid, camid). + transform: transform function. + k_tfm (int): number of times to apply augmentation to an image + independently. If k_tfm > 1, the transform function will be + applied k_tfm times to an image. This variable will only be + useful for training and is currently valid for image datasets only. + mode (str): 'train', 'query' or 'gallery'. + combineall (bool): combines train, query and gallery in a + dataset for training. + verbose (bool): show information. + """ + + # junk_pids contains useless person IDs, e.g. background, + # false detections, distractors. These IDs will be ignored + # when combining all images in a dataset for training, i.e. + # combineall=True + _junk_pids = [] + + # Some datasets are only used for training, like CUHK-SYSU + # In this case, "combineall=True" is not used for them + _train_only = False + + def __init__( + self, + train, + query, + gallery, + transform=None, + k_tfm=1, + mode='train', + combineall=False, + verbose=True, + **kwargs + ): + # extend 3-tuple (img_path(s), pid, camid) to + # 4-tuple (img_path(s), pid, camid, dsetid) by + # adding a dataset indicator "dsetid" + if len(train[0]) == 3: + train = [(*items, 0) for items in train] + if len(query[0]) == 3: + query = [(*items, 0) for items in query] + if len(gallery[0]) == 3: + gallery = [(*items, 0) for items in gallery] + + self.train = train + self.query = query + self.gallery = gallery + self.transform = transform + self.k_tfm = k_tfm + self.mode = mode + self.combineall = combineall + self.verbose = verbose + + self.num_train_pids = self.get_num_pids(self.train) + self.num_train_cams = self.get_num_cams(self.train) + self.num_datasets = self.get_num_datasets(self.train) + + if self.combineall: + self.combine_all() + + if self.mode == 'train': + self.data = self.train + elif self.mode == 'query': + self.data = self.query + elif self.mode == 'gallery': + self.data = self.gallery + else: + raise ValueError( + 'Invalid mode. Got {}, but expected to be ' + 'one of [train | query | gallery]'.format(self.mode) + ) + + if self.verbose: + self.show_summary() + + def __getitem__(self, index): + raise NotImplementedError + + def __len__(self): + return len(self.data) + + def __add__(self, other): + """Adds two datasets together (only the train set).""" + train = copy.deepcopy(self.train) + + for img_path, pid, camid, dsetid in other.train: + pid += self.num_train_pids + camid += self.num_train_cams + dsetid += self.num_datasets + train.append((img_path, pid, camid, dsetid)) + + ################################### + # Note that + # 1. set verbose=False to avoid unnecessary print + # 2. set combineall=False because combineall would have been applied + # if it was True for a specific dataset; setting it to True will + # create new IDs that should have already been included + ################################### + if isinstance(train[0][0], str): + return ImageDataset( + train, + self.query, + self.gallery, + transform=self.transform, + mode=self.mode, + combineall=False, + verbose=False + ) + else: + return VideoDataset( + train, + self.query, + self.gallery, + transform=self.transform, + mode=self.mode, + combineall=False, + verbose=False, + seq_len=self.seq_len, + sample_method=self.sample_method + ) + + def __radd__(self, other): + """Supports sum([dataset1, dataset2, dataset3]).""" + if other == 0: + return self + else: + return self.__add__(other) + + def get_num_pids(self, data): + """Returns the number of training person identities. + + Each tuple in data contains (img_path(s), pid, camid, dsetid). + """ + pids = set() + for items in data: + pid = items[1] + pids.add(pid) + return len(pids) + + def get_num_cams(self, data): + """Returns the number of training cameras. + + Each tuple in data contains (img_path(s), pid, camid, dsetid). + """ + cams = set() + for items in data: + camid = items[2] + cams.add(camid) + return len(cams) + + def get_num_datasets(self, data): + """Returns the number of datasets included. + + Each tuple in data contains (img_path(s), pid, camid, dsetid). + """ + dsets = set() + for items in data: + dsetid = items[3] + dsets.add(dsetid) + return len(dsets) + + def show_summary(self): + """Shows dataset statistics.""" + pass + + def combine_all(self): + """Combines train, query and gallery in a dataset for training.""" + if self._train_only: + return + + combined = copy.deepcopy(self.train) + + # relabel pids in gallery (query shares the same scope) + g_pids = set() + for items in self.gallery: + pid = items[1] + if pid in self._junk_pids: + continue + g_pids.add(pid) + pid2label = {pid: i for i, pid in enumerate(g_pids)} + + def _combine_data(data): + for img_path, pid, camid, dsetid in data: + if pid in self._junk_pids: + continue + pid = pid2label[pid] + self.num_train_pids + combined.append((img_path, pid, camid, dsetid)) + + _combine_data(self.query) + _combine_data(self.gallery) + + self.train = combined + self.num_train_pids = self.get_num_pids(self.train) + + def download_dataset(self, dataset_dir, dataset_url): + """Downloads and extracts dataset. + + Args: + dataset_dir (str): dataset directory. + dataset_url (str): url to download dataset. + """ + if osp.exists(dataset_dir): + return + + if dataset_url is None: + raise RuntimeError( + '{} dataset needs to be manually ' + 'prepared, please follow the ' + 'document to prepare this dataset'.format( + self.__class__.__name__ + ) + ) + + print('Creating directory "{}"'.format(dataset_dir)) + mkdir_if_missing(dataset_dir) + fpath = osp.join(dataset_dir, osp.basename(dataset_url)) + + print( + 'Downloading {} dataset to "{}"'.format( + self.__class__.__name__, dataset_dir + ) + ) + download_url(dataset_url, fpath) + + print('Extracting "{}"'.format(fpath)) + try: + tar = tarfile.open(fpath) + tar.extractall(path=dataset_dir) + tar.close() + except: + zip_ref = zipfile.ZipFile(fpath, 'r') + zip_ref.extractall(dataset_dir) + zip_ref.close() + + print('{} dataset is ready'.format(self.__class__.__name__)) + + def check_before_run(self, required_files): + """Checks if required files exist before going deeper. + + Args: + required_files (str or list): string file name(s). + """ + if isinstance(required_files, str): + required_files = [required_files] + + for fpath in required_files: + if not osp.exists(fpath): + raise RuntimeError('"{}" is not found'.format(fpath)) + + def __repr__(self): + num_train_pids = self.get_num_pids(self.train) + num_train_cams = self.get_num_cams(self.train) + + num_query_pids = self.get_num_pids(self.query) + num_query_cams = self.get_num_cams(self.query) + + num_gallery_pids = self.get_num_pids(self.gallery) + num_gallery_cams = self.get_num_cams(self.gallery) + + msg = ' ----------------------------------------\n' \ + ' subset | # ids | # items | # cameras\n' \ + ' ----------------------------------------\n' \ + ' train | {:5d} | {:7d} | {:9d}\n' \ + ' query | {:5d} | {:7d} | {:9d}\n' \ + ' gallery | {:5d} | {:7d} | {:9d}\n' \ + ' ----------------------------------------\n' \ + ' items: images/tracklets for image/video dataset\n'.format( + num_train_pids, len(self.train), num_train_cams, + num_query_pids, len(self.query), num_query_cams, + num_gallery_pids, len(self.gallery), num_gallery_cams + ) + + return msg + + def _transform_image(self, tfm, k_tfm, img0): + """Transforms a raw image (img0) k_tfm times with + the transform function tfm. + """ + img_list = [] + + for k in range(k_tfm): + img_list.append(tfm(img0)) + + img = img_list + if len(img) == 1: + img = img[0] + + return img + + +class ImageDataset(Dataset): + """A base class representing ImageDataset. + + All other image datasets should subclass it. + + ``__getitem__`` returns an image given index. + It will return ``img``, ``pid``, ``camid`` and ``img_path`` + where ``img`` has shape (channel, height, width). As a result, + data in each batch has shape (batch_size, channel, height, width). + """ + + def __init__(self, train, query, gallery, **kwargs): + super(ImageDataset, self).__init__(train, query, gallery, **kwargs) + + def __getitem__(self, index): + img_path, pid, camid, dsetid = self.data[index] + img = read_image(img_path) + if self.transform is not None: + img = self._transform_image(self.transform, self.k_tfm, img) + item = { + 'img': img, + 'pid': pid, + 'camid': camid, + 'impath': img_path, + 'dsetid': dsetid + } + return item + + def show_summary(self): + num_train_pids = self.get_num_pids(self.train) + num_train_cams = self.get_num_cams(self.train) + + num_query_pids = self.get_num_pids(self.query) + num_query_cams = self.get_num_cams(self.query) + + num_gallery_pids = self.get_num_pids(self.gallery) + num_gallery_cams = self.get_num_cams(self.gallery) + + print('=> Loaded {}'.format(self.__class__.__name__)) + print(' ----------------------------------------') + print(' subset | # ids | # images | # cameras') + print(' ----------------------------------------') + print( + ' train | {:5d} | {:8d} | {:9d}'.format( + num_train_pids, len(self.train), num_train_cams + ) + ) + print( + ' query | {:5d} | {:8d} | {:9d}'.format( + num_query_pids, len(self.query), num_query_cams + ) + ) + print( + ' gallery | {:5d} | {:8d} | {:9d}'.format( + num_gallery_pids, len(self.gallery), num_gallery_cams + ) + ) + print(' ----------------------------------------') + + +class VideoDataset(Dataset): + """A base class representing VideoDataset. + + All other video datasets should subclass it. + + ``__getitem__`` returns an image given index. + It will return ``imgs``, ``pid`` and ``camid`` + where ``imgs`` has shape (seq_len, channel, height, width). As a result, + data in each batch has shape (batch_size, seq_len, channel, height, width). + """ + + def __init__( + self, + train, + query, + gallery, + seq_len=15, + sample_method='evenly', + **kwargs + ): + super(VideoDataset, self).__init__(train, query, gallery, **kwargs) + self.seq_len = seq_len + self.sample_method = sample_method + + if self.transform is None: + raise RuntimeError('transform must not be None') + + def __getitem__(self, index): + img_paths, pid, camid, dsetid = self.data[index] + num_imgs = len(img_paths) + + if self.sample_method == 'random': + # Randomly samples seq_len images from a tracklet of length num_imgs, + # if num_imgs is smaller than seq_len, then replicates images + indices = np.arange(num_imgs) + replace = False if num_imgs >= self.seq_len else True + indices = np.random.choice( + indices, size=self.seq_len, replace=replace + ) + # sort indices to keep temporal order (comment it to be order-agnostic) + indices = np.sort(indices) + + elif self.sample_method == 'evenly': + # Evenly samples seq_len images from a tracklet + if num_imgs >= self.seq_len: + num_imgs -= num_imgs % self.seq_len + indices = np.arange(0, num_imgs, num_imgs / self.seq_len) + else: + # if num_imgs is smaller than seq_len, simply replicate the last image + # until the seq_len requirement is satisfied + indices = np.arange(0, num_imgs) + num_pads = self.seq_len - num_imgs + indices = np.concatenate( + [ + indices, + np.ones(num_pads).astype(np.int32) * (num_imgs-1) + ] + ) + assert len(indices) == self.seq_len + + elif self.sample_method == 'all': + # Samples all images in a tracklet. batch_size must be set to 1 + indices = np.arange(num_imgs) + + else: + raise ValueError( + 'Unknown sample method: {}'.format(self.sample_method) + ) + + imgs = [] + for index in indices: + img_path = img_paths[int(index)] + img = read_image(img_path) + if self.transform is not None: + img = self.transform(img) + img = img.unsqueeze(0) # img must be torch.Tensor + imgs.append(img) + imgs = torch.cat(imgs, dim=0) + + item = {'img': imgs, 'pid': pid, 'camid': camid, 'dsetid': dsetid} + + return item + + def show_summary(self): + num_train_pids = self.get_num_pids(self.train) + num_train_cams = self.get_num_cams(self.train) + + num_query_pids = self.get_num_pids(self.query) + num_query_cams = self.get_num_cams(self.query) + + num_gallery_pids = self.get_num_pids(self.gallery) + num_gallery_cams = self.get_num_cams(self.gallery) + + print('=> Loaded {}'.format(self.__class__.__name__)) + print(' -------------------------------------------') + print(' subset | # ids | # tracklets | # cameras') + print(' -------------------------------------------') + print( + ' train | {:5d} | {:11d} | {:9d}'.format( + num_train_pids, len(self.train), num_train_cams + ) + ) + print( + ' query | {:5d} | {:11d} | {:9d}'.format( + num_query_pids, len(self.query), num_query_cams + ) + ) + print( + ' gallery | {:5d} | {:11d} | {:9d}'.format( + num_gallery_pids, len(self.gallery), num_gallery_cams + ) + ) + print(' -------------------------------------------') diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py new file mode 100644 index 0000000000..a0ed9d5b0e --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/__init__.py @@ -0,0 +1,62 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .grid import GRID +from .prid import PRID +from .ilids import iLIDS +from .viper import VIPeR +from .cuhk01 import CUHK01 +from .cuhk02 import CUHK02 +from .cuhk03 import CUHK03 +from .msmt17 import MSMT17 +from .cuhksysu import CUHKSYSU +from .sensereid import SenseReID +from .market1501 import Market1501 +from .dukemtmcreid import DukeMTMCreID +from .university1652 import University1652 diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py new file mode 100644 index 0000000000..f10f400f8b --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk01.py @@ -0,0 +1,184 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import numpy as np +import os.path as osp +import zipfile + +from torchreid.utils import read_json, write_json + +from ..dataset import ImageDataset + + +class CUHK01(ImageDataset): + """CUHK01. + + Reference: + Li et al. Human Reidentification with Transferred Metric Learning. ACCV 2012. + + URL: ``_ + + Dataset statistics: + - identities: 971. + - images: 3884. + - cameras: 4. + + Note: CUHK01 and CUHK02 overlap. + """ + dataset_dir = 'cuhk01' + dataset_url = None + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.zip_path = osp.join(self.dataset_dir, 'CUHK01.zip') + self.campus_dir = osp.join(self.dataset_dir, 'campus') + self.split_path = osp.join(self.dataset_dir, 'splits.json') + + self.extract_file() + + required_files = [self.dataset_dir, self.campus_dir] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, but expected between 0 and {}' + .format(split_id, + len(splits) - 1) + ) + split = splits[split_id] + + train = split['train'] + query = split['query'] + gallery = split['gallery'] + + train = [tuple(item) for item in train] + query = [tuple(item) for item in query] + gallery = [tuple(item) for item in gallery] + + super(CUHK01, self).__init__(train, query, gallery, **kwargs) + + def extract_file(self): + if not osp.exists(self.campus_dir): + print('Extracting files') + zip_ref = zipfile.ZipFile(self.zip_path, 'r') + zip_ref.extractall(self.dataset_dir) + zip_ref.close() + + def prepare_split(self): + """ + Image name format: 0001001.png, where first four digits represent identity + and last four digits represent cameras. Camera 1&2 are considered the same + view and camera 3&4 are considered the same view. + """ + if not osp.exists(self.split_path): + print('Creating 10 random splits of train ids and test ids') + img_paths = sorted(glob.glob(osp.join(self.campus_dir, '*.png'))) + img_list = [] + pid_container = set() + for img_path in img_paths: + img_name = osp.basename(img_path) + pid = int(img_name[:4]) - 1 + camid = (int(img_name[4:7]) - 1) // 2 # result is either 0 or 1 + img_list.append((img_path, pid, camid)) + pid_container.add(pid) + + num_pids = len(pid_container) + num_train_pids = num_pids // 2 + + splits = [] + for _ in range(10): + order = np.arange(num_pids) + np.random.shuffle(order) + train_idxs = order[:num_train_pids] + train_idxs = np.sort(train_idxs) + idx2label = { + idx: label + for label, idx in enumerate(train_idxs) + } + + train, test_a, test_b = [], [], [] + for img_path, pid, camid in img_list: + if pid in train_idxs: + train.append((img_path, idx2label[pid], camid)) + else: + if camid == 0: + test_a.append((img_path, pid, camid)) + else: + test_b.append((img_path, pid, camid)) + + # use cameraA as query and cameraB as gallery + split = { + 'train': train, + 'query': test_a, + 'gallery': test_b, + 'num_train_pids': num_train_pids, + 'num_query_pids': num_pids - num_train_pids, + 'num_gallery_pids': num_pids - num_train_pids + } + splits.append(split) + + # use cameraB as query and cameraA as gallery + split = { + 'train': train, + 'query': test_b, + 'gallery': test_a, + 'num_train_pids': num_train_pids, + 'num_query_pids': num_pids - num_train_pids, + 'num_gallery_pids': num_pids - num_train_pids + } + splits.append(split) + + print('Totally {} splits are created'.format(len(splits))) + write_json(splits, self.split_path) + print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py new file mode 100644 index 0000000000..7eee2aa965 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk02.py @@ -0,0 +1,144 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import os.path as osp + +from ..dataset import ImageDataset + + +class CUHK02(ImageDataset): + """CUHK02. + + Reference: + Li and Wang. Locally Aligned Feature Transforms across Views. CVPR 2013. + + URL: ``_ + + Dataset statistics: + - 5 camera view pairs each with two cameras + - 971, 306, 107, 193 and 239 identities from P1 - P5 + - totally 1,816 identities + - image format is png + + Protocol: Use P1 - P4 for training and P5 for evaluation. + + Note: CUHK01 and CUHK02 overlap. + """ + dataset_dir = 'cuhk02' + cam_pairs = ['P1', 'P2', 'P3', 'P4', 'P5'] + test_cam_pair = 'P5' + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir, 'Dataset') + + required_files = [self.dataset_dir] + self.check_before_run(required_files) + + train, query, gallery = self.get_data_list() + + super(CUHK02, self).__init__(train, query, gallery, **kwargs) + + def get_data_list(self): + num_train_pids, camid = 0, 0 + train, query, gallery = [], [], [] + + for cam_pair in self.cam_pairs: + cam_pair_dir = osp.join(self.dataset_dir, cam_pair) + + cam1_dir = osp.join(cam_pair_dir, 'cam1') + cam2_dir = osp.join(cam_pair_dir, 'cam2') + + impaths1 = glob.glob(osp.join(cam1_dir, '*.png')) + impaths2 = glob.glob(osp.join(cam2_dir, '*.png')) + + if cam_pair == self.test_cam_pair: + # add images to query + for impath in impaths1: + pid = osp.basename(impath).split('_')[0] + pid = int(pid) + query.append((impath, pid, camid)) + camid += 1 + + # add images to gallery + for impath in impaths2: + pid = osp.basename(impath).split('_')[0] + pid = int(pid) + gallery.append((impath, pid, camid)) + camid += 1 + + else: + pids1 = [ + osp.basename(impath).split('_')[0] for impath in impaths1 + ] + pids2 = [ + osp.basename(impath).split('_')[0] for impath in impaths2 + ] + pids = set(pids1 + pids2) + pid2label = { + pid: label + num_train_pids + for label, pid in enumerate(pids) + } + + # add images to train from cam1 + for impath in impaths1: + pid = osp.basename(impath).split('_')[0] + pid = pid2label[pid] + train.append((impath, pid, camid)) + camid += 1 + + # add images to train from cam2 + for impath in impaths2: + pid = osp.basename(impath).split('_')[0] + pid = pid2label[pid] + train.append((impath, pid, camid)) + camid += 1 + num_train_pids += len(pids) + + return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py new file mode 100644 index 0000000000..8917929bdc --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhk03.py @@ -0,0 +1,354 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import os.path as osp + +from torchreid.utils import read_json, write_json, mkdir_if_missing + +from ..dataset import ImageDataset + + +class CUHK03(ImageDataset): + """CUHK03. + + Reference: + Li et al. DeepReID: Deep Filter Pairing Neural Network for Person Re-identification. CVPR 2014. + + URL: ``_ + + Dataset statistics: + - identities: 1360. + - images: 13164. + - cameras: 6. + - splits: 20 (classic). + """ + dataset_dir = 'cuhk03' + dataset_url = None + + def __init__( + self, + root='', + split_id=0, + cuhk03_labeled=False, + cuhk03_classic_split=False, + **kwargs + ): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.data_dir = osp.join(self.dataset_dir, 'cuhk03_release') + self.raw_mat_path = osp.join(self.data_dir, 'cuhk-03.mat') + + self.imgs_detected_dir = osp.join(self.dataset_dir, 'images_detected') + self.imgs_labeled_dir = osp.join(self.dataset_dir, 'images_labeled') + + self.split_classic_det_json_path = osp.join( + self.dataset_dir, 'splits_classic_detected.json' + ) + self.split_classic_lab_json_path = osp.join( + self.dataset_dir, 'splits_classic_labeled.json' + ) + + self.split_new_det_json_path = osp.join( + self.dataset_dir, 'splits_new_detected.json' + ) + self.split_new_lab_json_path = osp.join( + self.dataset_dir, 'splits_new_labeled.json' + ) + + self.split_new_det_mat_path = osp.join( + self.dataset_dir, 'cuhk03_new_protocol_config_detected.mat' + ) + self.split_new_lab_mat_path = osp.join( + self.dataset_dir, 'cuhk03_new_protocol_config_labeled.mat' + ) + + required_files = [ + self.dataset_dir, self.data_dir, self.raw_mat_path, + self.split_new_det_mat_path, self.split_new_lab_mat_path + ] + self.check_before_run(required_files) + + self.preprocess_split() + + if cuhk03_labeled: + split_path = self.split_classic_lab_json_path if cuhk03_classic_split else self.split_new_lab_json_path + else: + split_path = self.split_classic_det_json_path if cuhk03_classic_split else self.split_new_det_json_path + + splits = read_json(split_path) + assert split_id < len( + splits + ), 'Condition split_id ({}) < len(splits) ({}) is false'.format( + split_id, len(splits) + ) + split = splits[split_id] + + train = split['train'] + query = split['query'] + gallery = split['gallery'] + + super(CUHK03, self).__init__(train, query, gallery, **kwargs) + + def preprocess_split(self): + # This function is a bit complex and ugly, what it does is + # 1. extract data from cuhk-03.mat and save as png images + # 2. create 20 classic splits (Li et al. CVPR'14) + # 3. create new split (Zhong et al. CVPR'17) + if osp.exists(self.imgs_labeled_dir) \ + and osp.exists(self.imgs_detected_dir) \ + and osp.exists(self.split_classic_det_json_path) \ + and osp.exists(self.split_classic_lab_json_path) \ + and osp.exists(self.split_new_det_json_path) \ + and osp.exists(self.split_new_lab_json_path): + return + + import h5py + import imageio + from scipy.io import loadmat + + mkdir_if_missing(self.imgs_detected_dir) + mkdir_if_missing(self.imgs_labeled_dir) + + print( + 'Extract image data from "{}" and save as png'.format( + self.raw_mat_path + ) + ) + mat = h5py.File(self.raw_mat_path, 'r') + + def _deref(ref): + return mat[ref][:].T + + def _process_images(img_refs, campid, pid, save_dir): + img_paths = [] # Note: some persons only have images for one view + for imgid, img_ref in enumerate(img_refs): + img = _deref(img_ref) + if img.size == 0 or img.ndim < 3: + continue # skip empty cell + # images are saved with the following format, index-1 (ensure uniqueness) + # campid: index of camera pair (1-5) + # pid: index of person in 'campid'-th camera pair + # viewid: index of view, {1, 2} + # imgid: index of image, (1-10) + viewid = 1 if imgid < 5 else 2 + img_name = '{:01d}_{:03d}_{:01d}_{:02d}.png'.format( + campid + 1, pid + 1, viewid, imgid + 1 + ) + img_path = osp.join(save_dir, img_name) + if not osp.isfile(img_path): + imageio.imwrite(img_path, img) + img_paths.append(img_path) + return img_paths + + def _extract_img(image_type): + print('Processing {} images ...'.format(image_type)) + meta_data = [] + imgs_dir = self.imgs_detected_dir if image_type == 'detected' else self.imgs_labeled_dir + for campid, camp_ref in enumerate(mat[image_type][0]): + camp = _deref(camp_ref) + num_pids = camp.shape[0] + for pid in range(num_pids): + img_paths = _process_images( + camp[pid, :], campid, pid, imgs_dir + ) + assert len(img_paths) > 0, \ + 'campid{}-pid{} has no images'.format(campid, pid) + meta_data.append((campid + 1, pid + 1, img_paths)) + print( + '- done camera pair {} with {} identities'.format( + campid + 1, num_pids + ) + ) + return meta_data + + meta_detected = _extract_img('detected') + meta_labeled = _extract_img('labeled') + + def _extract_classic_split(meta_data, test_split): + train, test = [], [] + num_train_pids, num_test_pids = 0, 0 + num_train_imgs, num_test_imgs = 0, 0 + for i, (campid, pid, img_paths) in enumerate(meta_data): + + if [campid, pid] in test_split: + for img_path in img_paths: + camid = int( + osp.basename(img_path).split('_')[2] + ) - 1 # make it 0-based + test.append((img_path, num_test_pids, camid)) + num_test_pids += 1 + num_test_imgs += len(img_paths) + else: + for img_path in img_paths: + camid = int( + osp.basename(img_path).split('_')[2] + ) - 1 # make it 0-based + train.append((img_path, num_train_pids, camid)) + num_train_pids += 1 + num_train_imgs += len(img_paths) + return train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs + + print('Creating classic splits (# = 20) ...') + splits_classic_det, splits_classic_lab = [], [] + for split_ref in mat['testsets'][0]: + test_split = _deref(split_ref).tolist() + + # create split for detected images + train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs = \ + _extract_classic_split(meta_detected, test_split) + splits_classic_det.append( + { + 'train': train, + 'query': test, + 'gallery': test, + 'num_train_pids': num_train_pids, + 'num_train_imgs': num_train_imgs, + 'num_query_pids': num_test_pids, + 'num_query_imgs': num_test_imgs, + 'num_gallery_pids': num_test_pids, + 'num_gallery_imgs': num_test_imgs + } + ) + + # create split for labeled images + train, num_train_pids, num_train_imgs, test, num_test_pids, num_test_imgs = \ + _extract_classic_split(meta_labeled, test_split) + splits_classic_lab.append( + { + 'train': train, + 'query': test, + 'gallery': test, + 'num_train_pids': num_train_pids, + 'num_train_imgs': num_train_imgs, + 'num_query_pids': num_test_pids, + 'num_query_imgs': num_test_imgs, + 'num_gallery_pids': num_test_pids, + 'num_gallery_imgs': num_test_imgs + } + ) + + write_json(splits_classic_det, self.split_classic_det_json_path) + write_json(splits_classic_lab, self.split_classic_lab_json_path) + + def _extract_set(filelist, pids, pid2label, idxs, img_dir, relabel): + tmp_set = [] + unique_pids = set() + for idx in idxs: + img_name = filelist[idx][0] + camid = int(img_name.split('_')[2]) - 1 # make it 0-based + pid = pids[idx] + if relabel: + pid = pid2label[pid] + img_path = osp.join(img_dir, img_name) + tmp_set.append((img_path, int(pid), camid)) + unique_pids.add(pid) + return tmp_set, len(unique_pids), len(idxs) + + def _extract_new_split(split_dict, img_dir): + train_idxs = split_dict['train_idx'].flatten() - 1 # index-0 + pids = split_dict['labels'].flatten() + train_pids = set(pids[train_idxs]) + pid2label = {pid: label for label, pid in enumerate(train_pids)} + query_idxs = split_dict['query_idx'].flatten() - 1 + gallery_idxs = split_dict['gallery_idx'].flatten() - 1 + filelist = split_dict['filelist'].flatten() + train_info = _extract_set( + filelist, pids, pid2label, train_idxs, img_dir, relabel=True + ) + query_info = _extract_set( + filelist, pids, pid2label, query_idxs, img_dir, relabel=False + ) + gallery_info = _extract_set( + filelist, + pids, + pid2label, + gallery_idxs, + img_dir, + relabel=False + ) + return train_info, query_info, gallery_info + + print('Creating new split for detected images (767/700) ...') + train_info, query_info, gallery_info = _extract_new_split( + loadmat(self.split_new_det_mat_path), self.imgs_detected_dir + ) + split = [ + { + 'train': train_info[0], + 'query': query_info[0], + 'gallery': gallery_info[0], + 'num_train_pids': train_info[1], + 'num_train_imgs': train_info[2], + 'num_query_pids': query_info[1], + 'num_query_imgs': query_info[2], + 'num_gallery_pids': gallery_info[1], + 'num_gallery_imgs': gallery_info[2] + } + ] + write_json(split, self.split_new_det_json_path) + + print('Creating new split for labeled images (767/700) ...') + train_info, query_info, gallery_info = _extract_new_split( + loadmat(self.split_new_lab_mat_path), self.imgs_labeled_dir + ) + split = [ + { + 'train': train_info[0], + 'query': query_info[0], + 'gallery': gallery_info[0], + 'num_train_pids': train_info[1], + 'num_train_imgs': train_info[2], + 'num_query_pids': query_info[1], + 'num_query_imgs': query_info[2], + 'num_gallery_pids': gallery_info[1], + 'num_gallery_imgs': gallery_info[2] + } + ] + write_json(split, self.split_new_lab_json_path) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py new file mode 100644 index 0000000000..1e8e25977a --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/cuhksysu.py @@ -0,0 +1,107 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import copy +import glob +import os.path as osp + +from ..dataset import ImageDataset + + +class CUHKSYSU(ImageDataset): + """CUHKSYSU. + + This dataset can only be used for model training. + + Reference: + Xiao et al. End-to-end deep learning for person search. + + URL: ``_ + + Dataset statistics: + - identities: 11,934 + - images: 34,574 + """ + _train_only = True + dataset_dir = 'cuhksysu' + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.data_dir = osp.join(self.dataset_dir, 'cropped_images') + + # image name format: p11422_s16929_1.jpg + train = self.process_dir(self.data_dir) + query = [copy.deepcopy(train[0])] + gallery = [copy.deepcopy(train[0])] + + super(CUHKSYSU, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dirname): + img_paths = glob.glob(osp.join(dirname, '*.jpg')) + # num_imgs = len(img_paths) + + # get all identities: + pid_container = set() + for img_path in img_paths: + img_name = osp.basename(img_path) + pid = img_name.split('_')[0] + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + + # num_pids = len(pid_container) + + # extract data + data = [] + for img_path in img_paths: + img_name = osp.basename(img_path) + pid = img_name.split('_')[0] + label = pid2label[pid] + data.append((img_path, label, 0)) # dummy camera id + + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py new file mode 100644 index 0000000000..62ca860d6b --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/dukemtmcreid.py @@ -0,0 +1,115 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import re +import glob +import os.path as osp + +from ..dataset import ImageDataset + + +class DukeMTMCreID(ImageDataset): + """DukeMTMC-reID. + + Reference: + - Ristani et al. Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. ECCVW 2016. + - Zheng et al. Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro. ICCV 2017. + + URL: ``_ + + Dataset statistics: + - identities: 1404 (train + query). + - images:16522 (train) + 2228 (query) + 17661 (gallery). + - cameras: 8. + """ + dataset_dir = 'dukemtmc-reid' + dataset_url = 'http://vision.cs.duke.edu/DukeMTMC/data/misc/DukeMTMC-reID.zip' + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + self.train_dir = osp.join( + self.dataset_dir, 'DukeMTMC-reID/bounding_box_train' + ) + self.query_dir = osp.join(self.dataset_dir, 'DukeMTMC-reID/query') + self.gallery_dir = osp.join( + self.dataset_dir, 'DukeMTMC-reID/bounding_box_test' + ) + + required_files = [ + self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir + ] + self.check_before_run(required_files) + + train = self.process_dir(self.train_dir, relabel=True) + query = self.process_dir(self.query_dir, relabel=False) + gallery = self.process_dir(self.gallery_dir, relabel=False) + + super(DukeMTMCreID, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path, relabel=False): + img_paths = glob.glob(osp.join(dir_path, '*.jpg')) + pattern = re.compile(r'([-\d]+)_c(\d)') + + pid_container = set() + for img_path in img_paths: + pid, _ = map(int, pattern.search(img_path).groups()) + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + + data = [] + for img_path in img_paths: + pid, camid = map(int, pattern.search(img_path).groups()) + assert 1 <= camid <= 8 + camid -= 1 # index starts from 0 + if relabel: + pid = pid2label[pid] + data.append((img_path, pid, camid)) + + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py new file mode 100644 index 0000000000..bfe897b649 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/grid.py @@ -0,0 +1,178 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import os.path as osp +from scipy.io import loadmat + +from torchreid.utils import read_json, write_json + +from ..dataset import ImageDataset + + +class GRID(ImageDataset): + """GRID. + + Reference: + Loy et al. Multi-camera activity correlation analysis. CVPR 2009. + + URL: ``_ + + Dataset statistics: + - identities: 250. + - images: 1275. + - cameras: 8. + """ + dataset_dir = 'grid' + dataset_url = 'http://personal.ie.cuhk.edu.hk/~ccloy/files/datasets/underground_reid.zip' + _junk_pids = [0] + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.probe_path = osp.join( + self.dataset_dir, 'underground_reid', 'probe' + ) + self.gallery_path = osp.join( + self.dataset_dir, 'underground_reid', 'gallery' + ) + self.split_mat_path = osp.join( + self.dataset_dir, 'underground_reid', 'features_and_partitions.mat' + ) + self.split_path = osp.join(self.dataset_dir, 'splits.json') + + required_files = [ + self.dataset_dir, self.probe_path, self.gallery_path, + self.split_mat_path + ] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, ' + 'but expected between 0 and {}'.format( + split_id, + len(splits) - 1 + ) + ) + split = splits[split_id] + + train = split['train'] + query = split['query'] + gallery = split['gallery'] + + train = [tuple(item) for item in train] + query = [tuple(item) for item in query] + gallery = [tuple(item) for item in gallery] + + super(GRID, self).__init__(train, query, gallery, **kwargs) + + def prepare_split(self): + if not osp.exists(self.split_path): + print('Creating 10 random splits') + split_mat = loadmat(self.split_mat_path) + trainIdxAll = split_mat['trainIdxAll'][0] # length = 10 + probe_img_paths = sorted( + glob.glob(osp.join(self.probe_path, '*.jpeg')) + ) + gallery_img_paths = sorted( + glob.glob(osp.join(self.gallery_path, '*.jpeg')) + ) + + splits = [] + for split_idx in range(10): + train_idxs = trainIdxAll[split_idx][0][0][2][0].tolist() + assert len(train_idxs) == 125 + idx2label = { + idx: label + for label, idx in enumerate(train_idxs) + } + + train, query, gallery = [], [], [] + + # processing probe folder + for img_path in probe_img_paths: + img_name = osp.basename(img_path) + img_idx = int(img_name.split('_')[0]) + camid = int( + img_name.split('_')[1] + ) - 1 # index starts from 0 + if img_idx in train_idxs: + train.append((img_path, idx2label[img_idx], camid)) + else: + query.append((img_path, img_idx, camid)) + + # process gallery folder + for img_path in gallery_img_paths: + img_name = osp.basename(img_path) + img_idx = int(img_name.split('_')[0]) + camid = int( + img_name.split('_')[1] + ) - 1 # index starts from 0 + if img_idx in train_idxs: + train.append((img_path, idx2label[img_idx], camid)) + else: + gallery.append((img_path, img_idx, camid)) + + split = { + 'train': train, + 'query': query, + 'gallery': gallery, + 'num_train_pids': 125, + 'num_query_pids': 125, + 'num_gallery_pids': 900 + } + splits.append(split) + + print('Totally {} splits are created'.format(len(splits))) + write_json(splits, self.split_path) + print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py new file mode 100644 index 0000000000..e2a97265c3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/ilids.py @@ -0,0 +1,182 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import copy +import glob +import random +import os.path as osp +from collections import defaultdict + +from torchreid.utils import read_json, write_json + +from ..dataset import ImageDataset + + +class iLIDS(ImageDataset): + """QMUL-iLIDS. + + Reference: + Zheng et al. Associating Groups of People. BMVC 2009. + + Dataset statistics: + - identities: 119. + - images: 476. + - cameras: 8 (not explicitly provided). + """ + dataset_dir = 'ilids' + dataset_url = 'http://www.eecs.qmul.ac.uk/~jason/data/i-LIDS_Pedestrian.tgz' + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.data_dir = osp.join(self.dataset_dir, 'i-LIDS_Pedestrian/Persons') + self.split_path = osp.join(self.dataset_dir, 'splits.json') + + required_files = [self.dataset_dir, self.data_dir] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, but ' + 'expected between 0 and {}'.format(split_id, + len(splits) - 1) + ) + split = splits[split_id] + + train, query, gallery = self.process_split(split) + + super(iLIDS, self).__init__(train, query, gallery, **kwargs) + + def prepare_split(self): + if not osp.exists(self.split_path): + print('Creating splits ...') + + paths = glob.glob(osp.join(self.data_dir, '*.jpg')) + img_names = [osp.basename(path) for path in paths] + num_imgs = len(img_names) + assert num_imgs == 476, 'There should be 476 images, but ' \ + 'got {}, please check the data'.format(num_imgs) + + # store image names + # image naming format: + # the first four digits denote the person ID + # the last four digits denote the sequence index + pid_dict = defaultdict(list) + for img_name in img_names: + pid = int(img_name[:4]) + pid_dict[pid].append(img_name) + pids = list(pid_dict.keys()) + num_pids = len(pids) + assert num_pids == 119, 'There should be 119 identities, ' \ + 'but got {}, please check the data'.format(num_pids) + + num_train_pids = int(num_pids * 0.5) + + splits = [] + for _ in range(10): + # randomly choose num_train_pids train IDs and the rest for test IDs + pids_copy = copy.deepcopy(pids) + random.shuffle(pids_copy) + train_pids = pids_copy[:num_train_pids] + test_pids = pids_copy[num_train_pids:] + + train = [] + query = [] + gallery = [] + + # for train IDs, all images are used in the train set. + for pid in train_pids: + img_names = pid_dict[pid] + train.extend(img_names) + + # for each test ID, randomly choose two images, one for + # query and the other one for gallery. + for pid in test_pids: + img_names = pid_dict[pid] + samples = random.sample(img_names, 2) + query.append(samples[0]) + gallery.append(samples[1]) + + split = {'train': train, 'query': query, 'gallery': gallery} + splits.append(split) + + print('Totally {} splits are created'.format(len(splits))) + write_json(splits, self.split_path) + print('Split file is saved to {}'.format(self.split_path)) + + def get_pid2label(self, img_names): + pid_container = set() + for img_name in img_names: + pid = int(img_name[:4]) + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + return pid2label + + def parse_img_names(self, img_names, pid2label=None): + data = [] + + for img_name in img_names: + pid = int(img_name[:4]) + if pid2label is not None: + pid = pid2label[pid] + camid = int(img_name[4:7]) - 1 # 0-based + img_path = osp.join(self.data_dir, img_name) + data.append((img_path, pid, camid)) + + return data + + def process_split(self, split): + train_pid2label = self.get_pid2label(split['train']) + train = self.parse_img_names(split['train'], train_pid2label) + query = self.parse_img_names(split['query']) + gallery = self.parse_img_names(split['gallery']) + return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py new file mode 100644 index 0000000000..7cdf598f6e --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/market1501.py @@ -0,0 +1,133 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import re +import glob +import os.path as osp +import warnings + +from ..dataset import ImageDataset + + +class Market1501(ImageDataset): + """Market1501. + + Reference: + Zheng et al. Scalable Person Re-identification: A Benchmark. ICCV 2015. + + URL: ``_ + + Dataset statistics: + - identities: 1501 (+1 for background). + - images: 12936 (train) + 3368 (query) + 15913 (gallery). + """ + _junk_pids = [0, -1] + dataset_dir = 'market1501' + + def __init__(self, root='', market1501_500k=False, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + + # allow alternative directory structure + self.data_dir = self.dataset_dir + data_dir = osp.join(self.data_dir, 'Market-1501-v15.09.15') + if osp.isdir(data_dir): + self.data_dir = data_dir + else: + warnings.warn( + 'The current data structure is deprecated. Please ' + 'put data folders such as "bounding_box_train" under ' + '"Market-1501-v15.09.15".' + ) + + self.train_dir = osp.join(self.data_dir, 'bounding_box_train') + self.query_dir = osp.join(self.data_dir, 'query') + self.gallery_dir = osp.join(self.data_dir, 'bounding_box_test') + self.extra_gallery_dir = osp.join(self.data_dir, 'images') + self.market1501_500k = market1501_500k + + required_files = [ + self.data_dir, self.train_dir, self.query_dir, self.gallery_dir + ] + if self.market1501_500k: + required_files.append(self.extra_gallery_dir) + self.check_before_run(required_files) + + train = self.process_dir(self.train_dir, relabel=True) + query = self.process_dir(self.query_dir, relabel=False) + gallery = self.process_dir(self.gallery_dir, relabel=False) + if self.market1501_500k: + gallery += self.process_dir(self.extra_gallery_dir, relabel=False) + + super(Market1501, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path, relabel=False): + img_paths = glob.glob(osp.join(dir_path, '*.jpg')) + pattern = re.compile(r'([-\d]+)_c(\d)') + + pid_container = set() + for img_path in img_paths: + pid, _ = map(int, pattern.search(img_path).groups()) + if pid == -1: + continue # junk images are just ignored + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + + data = [] + for img_path in img_paths: + pid, camid = map(int, pattern.search(img_path).groups()) + if pid == -1: + continue # junk images are just ignored + assert 0 <= pid <= 1501 # pid == 0 means background + assert 1 <= camid <= 6 + camid -= 1 # index starts from 0 + if relabel: + pid = pid2label[pid] + data.append((img_path, pid, camid)) + + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py new file mode 100644 index 0000000000..6f5559bea3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/msmt17.py @@ -0,0 +1,145 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import os.path as osp + +from ..dataset import ImageDataset + +# Log +# 22.01.2019 +# - add v2 +# - v1 and v2 differ in dir names +# - note that faces in v2 are blurred +TRAIN_DIR_KEY = 'train_dir' +TEST_DIR_KEY = 'test_dir' +VERSION_DICT = { + 'MSMT17_V1': { + TRAIN_DIR_KEY: 'train', + TEST_DIR_KEY: 'test', + }, + 'MSMT17_V2': { + TRAIN_DIR_KEY: 'mask_train_v2', + TEST_DIR_KEY: 'mask_test_v2', + } +} + + +class MSMT17(ImageDataset): + """MSMT17. + + Reference: + Wei et al. Person Transfer GAN to Bridge Domain Gap for Person Re-Identification. CVPR 2018. + + URL: ``_ + + Dataset statistics: + - identities: 4101. + - images: 32621 (train) + 11659 (query) + 82161 (gallery). + - cameras: 15. + """ + dataset_dir = 'msmt17' + dataset_url = None + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + has_main_dir = False + for main_dir in VERSION_DICT: + if osp.exists(osp.join(self.dataset_dir, main_dir)): + train_dir = VERSION_DICT[main_dir][TRAIN_DIR_KEY] + test_dir = VERSION_DICT[main_dir][TEST_DIR_KEY] + has_main_dir = True + break + assert has_main_dir, 'Dataset folder not found' + + self.train_dir = osp.join(self.dataset_dir, main_dir, train_dir) + self.test_dir = osp.join(self.dataset_dir, main_dir, test_dir) + self.list_train_path = osp.join( + self.dataset_dir, main_dir, 'list_train.txt' + ) + self.list_val_path = osp.join( + self.dataset_dir, main_dir, 'list_val.txt' + ) + self.list_query_path = osp.join( + self.dataset_dir, main_dir, 'list_query.txt' + ) + self.list_gallery_path = osp.join( + self.dataset_dir, main_dir, 'list_gallery.txt' + ) + + required_files = [self.dataset_dir, self.train_dir, self.test_dir] + self.check_before_run(required_files) + + train = self.process_dir(self.train_dir, self.list_train_path) + val = self.process_dir(self.train_dir, self.list_val_path) + query = self.process_dir(self.test_dir, self.list_query_path) + gallery = self.process_dir(self.test_dir, self.list_gallery_path) + + # Note: to fairly compare with published methods on the conventional ReID setting, + # do not add val images to the training set. + if 'combineall' in kwargs and kwargs['combineall']: + train += val + + super(MSMT17, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path, list_path): + with open(list_path, 'r') as txt: + lines = txt.readlines() + + data = [] + + for img_idx, img_info in enumerate(lines): + img_path, pid = img_info.split(' ') + pid = int(pid) # no need to relabel + camid = int(img_path.split('_')[2]) - 1 # index starts from 0 + img_path = osp.join(dir_path, img_path) + data.append((img_path, pid, camid)) + + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py new file mode 100644 index 0000000000..cebb77d1b9 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/prid.py @@ -0,0 +1,154 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import random +import os.path as osp + +from torchreid.utils import read_json, write_json + +from ..dataset import ImageDataset + + +class PRID(ImageDataset): + """PRID (single-shot version of prid-2011) + + Reference: + Hirzer et al. Person Re-Identification by Descriptive and Discriminative + Classification. SCIA 2011. + + URL: ``_ + + Dataset statistics: + - Two views. + - View A captures 385 identities. + - View B captures 749 identities. + - 200 identities appear in both views (index starts from 1 to 200). + """ + dataset_dir = 'prid2011' + dataset_url = None + _junk_pids = list(range(201, 750)) + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.cam_a_dir = osp.join( + self.dataset_dir, 'prid_2011', 'single_shot', 'cam_a' + ) + self.cam_b_dir = osp.join( + self.dataset_dir, 'prid_2011', 'single_shot', 'cam_b' + ) + self.split_path = osp.join(self.dataset_dir, 'splits_single_shot.json') + + required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, but expected between 0 and {}' + .format(split_id, + len(splits) - 1) + ) + split = splits[split_id] + + train, query, gallery = self.process_split(split) + + super(PRID, self).__init__(train, query, gallery, **kwargs) + + def prepare_split(self): + if not osp.exists(self.split_path): + print('Creating splits ...') + + splits = [] + for _ in range(10): + # randomly sample 100 IDs for train and use the rest 100 IDs for test + # (note: there are only 200 IDs appearing in both views) + pids = [i for i in range(1, 201)] + train_pids = random.sample(pids, 100) + train_pids.sort() + test_pids = [i for i in pids if i not in train_pids] + split = {'train': train_pids, 'test': test_pids} + splits.append(split) + + print('Totally {} splits are created'.format(len(splits))) + write_json(splits, self.split_path) + print('Split file is saved to {}'.format(self.split_path)) + + def process_split(self, split): + train_pids = split['train'] + test_pids = split['test'] + + train_pid2label = {pid: label for label, pid in enumerate(train_pids)} + + # train + train = [] + for pid in train_pids: + img_name = 'person_' + str(pid).zfill(4) + '.png' + pid = train_pid2label[pid] + img_a_path = osp.join(self.cam_a_dir, img_name) + train.append((img_a_path, pid, 0)) + img_b_path = osp.join(self.cam_b_dir, img_name) + train.append((img_b_path, pid, 1)) + + # query and gallery + query, gallery = [], [] + for pid in test_pids: + img_name = 'person_' + str(pid).zfill(4) + '.png' + img_a_path = osp.join(self.cam_a_dir, img_name) + query.append((img_a_path, pid, 0)) + img_b_path = osp.join(self.cam_b_dir, img_name) + gallery.append((img_b_path, pid, 1)) + for pid in range(201, 750): + img_name = 'person_' + str(pid).zfill(4) + '.png' + img_b_path = osp.join(self.cam_b_dir, img_name) + gallery.append((img_b_path, pid, 1)) + + return train, query, gallery diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py new file mode 100644 index 0000000000..22547885c6 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/sensereid.py @@ -0,0 +1,117 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import copy +import glob +import os.path as osp + +from ..dataset import ImageDataset + + +class SenseReID(ImageDataset): + """SenseReID. + + This dataset is used for test purpose only. + + Reference: + Zhao et al. Spindle Net: Person Re-identification with Human Body + Region Guided Feature Decomposition and Fusion. CVPR 2017. + + URL: ``_ + + Dataset statistics: + - query: 522 ids, 1040 images. + - gallery: 1717 ids, 3388 images. + """ + dataset_dir = 'sensereid' + dataset_url = None + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.query_dir = osp.join(self.dataset_dir, 'SenseReID', 'test_probe') + self.gallery_dir = osp.join( + self.dataset_dir, 'SenseReID', 'test_gallery' + ) + + required_files = [self.dataset_dir, self.query_dir, self.gallery_dir] + self.check_before_run(required_files) + + query = self.process_dir(self.query_dir) + gallery = self.process_dir(self.gallery_dir) + + # relabel + g_pids = set() + for _, pid, _ in gallery: + g_pids.add(pid) + pid2label = {pid: i for i, pid in enumerate(g_pids)} + + query = [ + (img_path, pid2label[pid], camid) for img_path, pid, camid in query + ] + gallery = [ + (img_path, pid2label[pid], camid) + for img_path, pid, camid in gallery + ] + train = copy.deepcopy(query) + copy.deepcopy(gallery) # dummy variable + + super(SenseReID, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path): + img_paths = glob.glob(osp.join(dir_path, '*.jpg')) + data = [] + + for img_path in img_paths: + img_name = osp.splitext(osp.basename(img_path))[0] + pid, camid = img_name.split('_') + pid, camid = int(pid), int(camid) + data.append((img_path, pid, camid)) + + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py new file mode 100644 index 0000000000..bd950fe3d8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/university1652.py @@ -0,0 +1,157 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import os +import glob +import os.path as osp +import gdown + +from ..dataset import ImageDataset + + +class University1652(ImageDataset): + """University-1652. + + Reference: + - Zheng et al. University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization. ACM MM 2020. + + URL: ``_ + OneDrive: + https://studentutsedu-my.sharepoint.com/:u:/g/personal/12639605_student_uts_edu_au/Ecrz6xK-PcdCjFdpNb0T0s8B_9J5ynaUy3q63_XumjJyrA?e=z4hpcz + [Backup] GoogleDrive: + https://drive.google.com/file/d/1iVnP4gjw-iHXa0KerZQ1IfIO0i1jADsR/view?usp=sharing + [Backup] Baidu Yun: + https://pan.baidu.com/s/1H_wBnWwikKbaBY1pMPjoqQ password: hrqp + + Dataset statistics: + - buildings: 1652 (train + query). + - The dataset split is as follows: + | Split | #imgs | #buildings | #universities| + | -------- | ----- | ----| ----| + | Training | 50,218 | 701 | 33 | + | Query_drone | 37,855 | 701 | 39 | + | Query_satellite | 701 | 701 | 39| + | Query_ground | 2,579 | 701 | 39| + | Gallery_drone | 51,355 | 951 | 39| + | Gallery_satellite | 951 | 951 | 39| + | Gallery_ground | 2,921 | 793 | 39| + - cameras: None. + + datamanager = torchreid.data.ImageDataManager( + root='reid-data', + sources='university1652', + targets='university1652', + height=256, + width=256, + batch_size_train=32, + batch_size_test=100, + transforms=['random_flip', 'random_crop'] + ) + """ + dataset_dir = 'university1652' + dataset_url = 'https://drive.google.com/uc?id=1iVnP4gjw-iHXa0KerZQ1IfIO0i1jADsR' + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + print(self.dataset_dir) + if not os.path.isdir(self.dataset_dir): + os.mkdir(self.dataset_dir) + gdown.download( + self.dataset_url, self.dataset_dir + 'data.zip', quiet=False + ) + os.system('unzip %s' % (self.dataset_dir + 'data.zip')) + self.train_dir = osp.join( + self.dataset_dir, 'University-Release/train/' + ) + self.query_dir = osp.join( + self.dataset_dir, 'University-Release/test/query_drone' + ) + self.gallery_dir = osp.join( + self.dataset_dir, 'University-Release/test/gallery_satellite' + ) + + required_files = [ + self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir + ] + self.check_before_run(required_files) + + self.fake_camid = 0 + train = self.process_dir(self.train_dir, relabel=True, train=True) + query = self.process_dir(self.query_dir, relabel=False) + gallery = self.process_dir(self.gallery_dir, relabel=False) + + super(University1652, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path, relabel=False, train=False): + IMG_EXTENSIONS = ( + '.jpg', '.jpeg', '.png', '.ppm', '.bmp', '.pgm', '.tif', '.tiff', + '.webp' + ) + if train: + img_paths = glob.glob(osp.join(dir_path, '*/*/*')) + else: + img_paths = glob.glob(osp.join(dir_path, '*/*')) + pid_container = set() + for img_path in img_paths: + if not img_path.lower().endswith(IMG_EXTENSIONS): + continue + pid = int(os.path.basename(os.path.dirname(img_path))) + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + data = [] + # no camera for university + for img_path in img_paths: + if not img_path.lower().endswith(IMG_EXTENSIONS): + continue + pid = int(os.path.basename(os.path.dirname(img_path))) + if relabel: + pid = pid2label[pid] + data.append((img_path, pid, self.fake_camid)) + self.fake_camid += 1 + return data diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py new file mode 100644 index 0000000000..efc64c1460 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/image/viper.py @@ -0,0 +1,175 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import numpy as np +import os.path as osp + +from torchreid.utils import read_json, write_json + +from ..dataset import ImageDataset + + +class VIPeR(ImageDataset): + """VIPeR. + + Reference: + Gray et al. Evaluating appearance models for recognition, reacquisition, and tracking. PETS 2007. + + URL: ``_ + + Dataset statistics: + - identities: 632. + - images: 632 x 2 = 1264. + - cameras: 2. + """ + dataset_dir = 'viper' + dataset_url = 'http://users.soe.ucsc.edu/~manduchi/VIPeR.v1.0.zip' + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.cam_a_dir = osp.join(self.dataset_dir, 'VIPeR', 'cam_a') + self.cam_b_dir = osp.join(self.dataset_dir, 'VIPeR', 'cam_b') + self.split_path = osp.join(self.dataset_dir, 'splits.json') + + required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, ' + 'but expected between 0 and {}'.format( + split_id, + len(splits) - 1 + ) + ) + split = splits[split_id] + + train = split['train'] + query = split['query'] # query and gallery share the same images + gallery = split['gallery'] + + train = [tuple(item) for item in train] + query = [tuple(item) for item in query] + gallery = [tuple(item) for item in gallery] + + super(VIPeR, self).__init__(train, query, gallery, **kwargs) + + def prepare_split(self): + if not osp.exists(self.split_path): + print('Creating 10 random splits of train ids and test ids') + + cam_a_imgs = sorted(glob.glob(osp.join(self.cam_a_dir, '*.bmp'))) + cam_b_imgs = sorted(glob.glob(osp.join(self.cam_b_dir, '*.bmp'))) + assert len(cam_a_imgs) == len(cam_b_imgs) + num_pids = len(cam_a_imgs) + print('Number of identities: {}'.format(num_pids)) + num_train_pids = num_pids // 2 + """ + In total, there will be 20 splits because each random split creates two + sub-splits, one using cameraA as query and cameraB as gallery + while the other using cameraB as query and cameraA as gallery. + Therefore, results should be averaged over 20 splits (split_id=0~19). + + In practice, a model trained on split_id=0 can be applied to split_id=0&1 + as split_id=0&1 share the same training data (so on and so forth). + """ + splits = [] + for _ in range(10): + order = np.arange(num_pids) + np.random.shuffle(order) + train_idxs = order[:num_train_pids] + test_idxs = order[num_train_pids:] + assert not bool(set(train_idxs) & set(test_idxs)), \ + 'Error: train and test overlap' + + train = [] + for pid, idx in enumerate(train_idxs): + cam_a_img = cam_a_imgs[idx] + cam_b_img = cam_b_imgs[idx] + train.append((cam_a_img, pid, 0)) + train.append((cam_b_img, pid, 1)) + + test_a = [] + test_b = [] + for pid, idx in enumerate(test_idxs): + cam_a_img = cam_a_imgs[idx] + cam_b_img = cam_b_imgs[idx] + test_a.append((cam_a_img, pid, 0)) + test_b.append((cam_b_img, pid, 1)) + + # use cameraA as query and cameraB as gallery + split = { + 'train': train, + 'query': test_a, + 'gallery': test_b, + 'num_train_pids': num_train_pids, + 'num_query_pids': num_pids - num_train_pids, + 'num_gallery_pids': num_pids - num_train_pids + } + splits.append(split) + + # use cameraB as query and cameraA as gallery + split = { + 'train': train, + 'query': test_b, + 'gallery': test_a, + 'num_train_pids': num_train_pids, + 'num_query_pids': num_pids - num_train_pids, + 'num_gallery_pids': num_pids - num_train_pids + } + splits.append(split) + + print('Totally {} splits are created'.format(len(splits))) + write_json(splits, self.split_path) + print('Split file saved to {}'.format(self.split_path)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py new file mode 100644 index 0000000000..141e7110c1 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/__init__.py @@ -0,0 +1,53 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .mars import Mars +from .ilidsvid import iLIDSVID +from .prid2011 import PRID2011 +from .dukemtmcvidreid import DukeMTMCVidReID diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py new file mode 100644 index 0000000000..895ff0b38f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/dukemtmcvidreid.py @@ -0,0 +1,175 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import os.path as osp +import warnings + +from torchreid.utils import read_json, write_json + +from ..dataset import VideoDataset + + +class DukeMTMCVidReID(VideoDataset): + """DukeMTMCVidReID. + + Reference: + - Ristani et al. Performance Measures and a Data Set for Multi-Target, + Multi-Camera Tracking. ECCVW 2016. + - Wu et al. Exploit the Unknown Gradually: One-Shot Video-Based Person + Re-Identification by Stepwise Learning. CVPR 2018. + + URL: ``_ + + Dataset statistics: + - identities: 702 (train) + 702 (test). + - tracklets: 2196 (train) + 2636 (test). + """ + dataset_dir = 'dukemtmc-vidreid' + dataset_url = 'http://vision.cs.duke.edu/DukeMTMC/data/misc/DukeMTMC-VideoReID.zip' + + def __init__(self, root='', min_seq_len=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.train_dir = osp.join(self.dataset_dir, 'DukeMTMC-VideoReID/train') + self.query_dir = osp.join(self.dataset_dir, 'DukeMTMC-VideoReID/query') + self.gallery_dir = osp.join( + self.dataset_dir, 'DukeMTMC-VideoReID/gallery' + ) + self.split_train_json_path = osp.join( + self.dataset_dir, 'split_train.json' + ) + self.split_query_json_path = osp.join( + self.dataset_dir, 'split_query.json' + ) + self.split_gallery_json_path = osp.join( + self.dataset_dir, 'split_gallery.json' + ) + self.min_seq_len = min_seq_len + + required_files = [ + self.dataset_dir, self.train_dir, self.query_dir, self.gallery_dir + ] + self.check_before_run(required_files) + + train = self.process_dir( + self.train_dir, self.split_train_json_path, relabel=True + ) + query = self.process_dir( + self.query_dir, self.split_query_json_path, relabel=False + ) + gallery = self.process_dir( + self.gallery_dir, self.split_gallery_json_path, relabel=False + ) + + super(DukeMTMCVidReID, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dir_path, json_path, relabel): + if osp.exists(json_path): + split = read_json(json_path) + return split['tracklets'] + + print('=> Generating split json file (** this might take a while **)') + pdirs = glob.glob(osp.join(dir_path, '*')) # avoid .DS_Store + print( + 'Processing "{}" with {} person identities'.format( + dir_path, len(pdirs) + ) + ) + + pid_container = set() + for pdir in pdirs: + pid = int(osp.basename(pdir)) + pid_container.add(pid) + pid2label = {pid: label for label, pid in enumerate(pid_container)} + + tracklets = [] + for pdir in pdirs: + pid = int(osp.basename(pdir)) + if relabel: + pid = pid2label[pid] + tdirs = glob.glob(osp.join(pdir, '*')) + for tdir in tdirs: + raw_img_paths = glob.glob(osp.join(tdir, '*.jpg')) + num_imgs = len(raw_img_paths) + + if num_imgs < self.min_seq_len: + continue + + img_paths = [] + for img_idx in range(num_imgs): + # some tracklet starts from 0002 instead of 0001 + img_idx_name = 'F' + str(img_idx + 1).zfill(4) + res = glob.glob( + osp.join(tdir, '*' + img_idx_name + '*.jpg') + ) + if len(res) == 0: + warnings.warn( + 'Index name {} in {} is missing, skip'.format( + img_idx_name, tdir + ) + ) + continue + img_paths.append(res[0]) + img_name = osp.basename(img_paths[0]) + if img_name.find('_') == -1: + # old naming format: 0001C6F0099X30823.jpg + camid = int(img_name[5]) - 1 + else: + # new naming format: 0001_C6_F0099_X30823.jpg + camid = int(img_name[6]) - 1 + img_paths = tuple(img_paths) + tracklets.append((img_paths, pid, camid)) + + print('Saving split to {}'.format(json_path)) + split_dict = {'tracklets': tracklets} + write_json(split_dict, json_path) + + return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py new file mode 100644 index 0000000000..7e61913d2b --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/ilidsvid.py @@ -0,0 +1,190 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import os.path as osp +from scipy.io import loadmat + +from torchreid.utils import read_json, write_json + +from ..dataset import VideoDataset + + +class iLIDSVID(VideoDataset): + """iLIDS-VID. + + Reference: + Wang et al. Person Re-Identification by Video Ranking. ECCV 2014. + + URL: ``_ + + Dataset statistics: + - identities: 300. + - tracklets: 600. + - cameras: 2. + """ + dataset_dir = 'ilids-vid' + dataset_url = 'http://www.eecs.qmul.ac.uk/~xiatian/iLIDS-VID/iLIDS-VID.tar' + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.data_dir = osp.join(self.dataset_dir, 'i-LIDS-VID') + self.split_dir = osp.join(self.dataset_dir, 'train-test people splits') + self.split_mat_path = osp.join( + self.split_dir, 'train_test_splits_ilidsvid.mat' + ) + self.split_path = osp.join(self.dataset_dir, 'splits.json') + self.cam_1_path = osp.join( + self.dataset_dir, 'i-LIDS-VID/sequences/cam1' + ) + self.cam_2_path = osp.join( + self.dataset_dir, 'i-LIDS-VID/sequences/cam2' + ) + + required_files = [self.dataset_dir, self.data_dir, self.split_dir] + self.check_before_run(required_files) + + self.prepare_split() + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, but expected between 0 and {}' + .format(split_id, + len(splits) - 1) + ) + split = splits[split_id] + train_dirs, test_dirs = split['train'], split['test'] + + train = self.process_data(train_dirs, cam1=True, cam2=True) + query = self.process_data(test_dirs, cam1=True, cam2=False) + gallery = self.process_data(test_dirs, cam1=False, cam2=True) + + super(iLIDSVID, self).__init__(train, query, gallery, **kwargs) + + def prepare_split(self): + if not osp.exists(self.split_path): + print('Creating splits ...') + mat_split_data = loadmat(self.split_mat_path)['ls_set'] + + num_splits = mat_split_data.shape[0] + num_total_ids = mat_split_data.shape[1] + assert num_splits == 10 + assert num_total_ids == 300 + num_ids_each = num_total_ids // 2 + + # pids in mat_split_data are indices, so we need to transform them + # to real pids + person_cam1_dirs = sorted( + glob.glob(osp.join(self.cam_1_path, '*')) + ) + person_cam2_dirs = sorted( + glob.glob(osp.join(self.cam_2_path, '*')) + ) + + person_cam1_dirs = [ + osp.basename(item) for item in person_cam1_dirs + ] + person_cam2_dirs = [ + osp.basename(item) for item in person_cam2_dirs + ] + + # make sure persons in one camera view can be found in the other camera view + assert set(person_cam1_dirs) == set(person_cam2_dirs) + + splits = [] + for i_split in range(num_splits): + # first 50% for testing and the remaining for training, following Wang et al. ECCV'14. + train_idxs = sorted( + list(mat_split_data[i_split, num_ids_each:]) + ) + test_idxs = sorted( + list(mat_split_data[i_split, :num_ids_each]) + ) + + train_idxs = [int(i) - 1 for i in train_idxs] + test_idxs = [int(i) - 1 for i in test_idxs] + + # transform pids to person dir names + train_dirs = [person_cam1_dirs[i] for i in train_idxs] + test_dirs = [person_cam1_dirs[i] for i in test_idxs] + + split = {'train': train_dirs, 'test': test_dirs} + splits.append(split) + + print( + 'Totally {} splits are created, following Wang et al. ECCV\'14' + .format(len(splits)) + ) + print('Split file is saved to {}'.format(self.split_path)) + write_json(splits, self.split_path) + + def process_data(self, dirnames, cam1=True, cam2=True): + tracklets = [] + dirname2pid = {dirname: i for i, dirname in enumerate(dirnames)} + + for dirname in dirnames: + if cam1: + person_dir = osp.join(self.cam_1_path, dirname) + img_names = glob.glob(osp.join(person_dir, '*.png')) + assert len(img_names) > 0 + img_names = tuple(img_names) + pid = dirname2pid[dirname] + tracklets.append((img_names, pid, 0)) + + if cam2: + person_dir = osp.join(self.cam_2_path, dirname) + img_names = glob.glob(osp.join(person_dir, '*.png')) + assert len(img_names) > 0 + img_names = tuple(img_names) + pid = dirname2pid[dirname] + tracklets.append((img_names, pid, 1)) + + return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py new file mode 100644 index 0000000000..fb6d215668 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/mars.py @@ -0,0 +1,180 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import os.path as osp +import warnings +from scipy.io import loadmat + +from ..dataset import VideoDataset + + +class Mars(VideoDataset): + """MARS. + + Reference: + Zheng et al. MARS: A Video Benchmark for Large-Scale Person Re-identification. ECCV 2016. + + URL: ``_ + + Dataset statistics: + - identities: 1261. + - tracklets: 8298 (train) + 1980 (query) + 9330 (gallery). + - cameras: 6. + """ + dataset_dir = 'mars' + dataset_url = None + + def __init__(self, root='', **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.train_name_path = osp.join( + self.dataset_dir, 'info/train_name.txt' + ) + self.test_name_path = osp.join(self.dataset_dir, 'info/test_name.txt') + self.track_train_info_path = osp.join( + self.dataset_dir, 'info/tracks_train_info.mat' + ) + self.track_test_info_path = osp.join( + self.dataset_dir, 'info/tracks_test_info.mat' + ) + self.query_IDX_path = osp.join(self.dataset_dir, 'info/query_IDX.mat') + + required_files = [ + self.dataset_dir, self.train_name_path, self.test_name_path, + self.track_train_info_path, self.track_test_info_path, + self.query_IDX_path + ] + self.check_before_run(required_files) + + train_names = self.get_names(self.train_name_path) + test_names = self.get_names(self.test_name_path) + track_train = loadmat(self.track_train_info_path + )['track_train_info'] # numpy.ndarray (8298, 4) + track_test = loadmat(self.track_test_info_path + )['track_test_info'] # numpy.ndarray (12180, 4) + query_IDX = loadmat(self.query_IDX_path + )['query_IDX'].squeeze() # numpy.ndarray (1980,) + query_IDX -= 1 # index from 0 + track_query = track_test[query_IDX, :] + gallery_IDX = [ + i for i in range(track_test.shape[0]) if i not in query_IDX + ] + track_gallery = track_test[gallery_IDX, :] + + train = self.process_data( + train_names, track_train, home_dir='bbox_train', relabel=True + ) + query = self.process_data( + test_names, track_query, home_dir='bbox_test', relabel=False + ) + gallery = self.process_data( + test_names, track_gallery, home_dir='bbox_test', relabel=False + ) + + super(Mars, self).__init__(train, query, gallery, **kwargs) + + def get_names(self, fpath): + names = [] + with open(fpath, 'r') as f: + for line in f: + new_line = line.rstrip() + names.append(new_line) + return names + + def process_data( + self, names, meta_data, home_dir=None, relabel=False, min_seq_len=0 + ): + assert home_dir in ['bbox_train', 'bbox_test'] + num_tracklets = meta_data.shape[0] + pid_list = list(set(meta_data[:, 2].tolist())) + + if relabel: + pid2label = {pid: label for label, pid in enumerate(pid_list)} + tracklets = [] + + for tracklet_idx in range(num_tracklets): + data = meta_data[tracklet_idx, ...] + start_index, end_index, pid, camid = data + if pid == -1: + continue # junk images are just ignored + assert 1 <= camid <= 6 + if relabel: + pid = pid2label[pid] + camid -= 1 # index starts from 0 + img_names = names[start_index - 1:end_index] + + # make sure image names correspond to the same person + pnames = [img_name[:4] for img_name in img_names] + assert len( + set(pnames) + ) == 1, 'Error: a single tracklet contains different person images' + + # make sure all images are captured under the same camera + camnames = [img_name[5] for img_name in img_names] + assert len( + set(camnames) + ) == 1, 'Error: images are captured under different cameras!' + + # append image names with directory information + img_paths = [ + osp.join(self.dataset_dir, home_dir, img_name[:4], img_name) + for img_name in img_names + ] + if len(img_paths) >= min_seq_len: + img_paths = tuple(img_paths) + tracklets.append((img_paths, pid, camid)) + + return tracklets + + def combine_all(self): + warnings.warn( + 'Some query IDs do not appear in gallery. Therefore, combineall ' + 'does not make any difference to Mars' + ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py new file mode 100644 index 0000000000..89598b85a9 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/datasets/video/prid2011.py @@ -0,0 +1,127 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import glob +import os.path as osp + +from torchreid.utils import read_json + +from ..dataset import VideoDataset + + +class PRID2011(VideoDataset): + """PRID2011. + + Reference: + Hirzer et al. Person Re-Identification by Descriptive and + Discriminative Classification. SCIA 2011. + + URL: ``_ + + Dataset statistics: + - identities: 200. + - tracklets: 400. + - cameras: 2. + """ + dataset_dir = 'prid2011' + dataset_url = None + + def __init__(self, root='', split_id=0, **kwargs): + self.root = osp.abspath(osp.expanduser(root)) + self.dataset_dir = osp.join(self.root, self.dataset_dir) + self.download_dataset(self.dataset_dir, self.dataset_url) + + self.split_path = osp.join(self.dataset_dir, 'splits_prid2011.json') + self.cam_a_dir = osp.join( + self.dataset_dir, 'prid_2011', 'multi_shot', 'cam_a' + ) + self.cam_b_dir = osp.join( + self.dataset_dir, 'prid_2011', 'multi_shot', 'cam_b' + ) + + required_files = [self.dataset_dir, self.cam_a_dir, self.cam_b_dir] + self.check_before_run(required_files) + + splits = read_json(self.split_path) + if split_id >= len(splits): + raise ValueError( + 'split_id exceeds range, received {}, but expected between 0 and {}' + .format(split_id, + len(splits) - 1) + ) + split = splits[split_id] + train_dirs, test_dirs = split['train'], split['test'] + + train = self.process_dir(train_dirs, cam1=True, cam2=True) + query = self.process_dir(test_dirs, cam1=True, cam2=False) + gallery = self.process_dir(test_dirs, cam1=False, cam2=True) + + super(PRID2011, self).__init__(train, query, gallery, **kwargs) + + def process_dir(self, dirnames, cam1=True, cam2=True): + tracklets = [] + dirname2pid = {dirname: i for i, dirname in enumerate(dirnames)} + + for dirname in dirnames: + if cam1: + person_dir = osp.join(self.cam_a_dir, dirname) + img_names = glob.glob(osp.join(person_dir, '*.png')) + assert len(img_names) > 0 + img_names = tuple(img_names) + pid = dirname2pid[dirname] + tracklets.append((img_names, pid, 0)) + + if cam2: + person_dir = osp.join(self.cam_b_dir, dirname) + img_names = glob.glob(osp.join(person_dir, '*.png')) + assert len(img_names) > 0 + img_names = tuple(img_names) + pid = dirname2pid[dirname] + tracklets.append((img_names, pid, 1)) + + return tracklets diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py new file mode 100644 index 0000000000..daf0d026c3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/sampler.py @@ -0,0 +1,292 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import copy +import numpy as np +import random +from collections import defaultdict +from torch.utils.data.sampler import Sampler, RandomSampler, SequentialSampler + +AVAI_SAMPLERS = [ + 'RandomIdentitySampler', 'SequentialSampler', 'RandomSampler', + 'RandomDomainSampler', 'RandomDatasetSampler' +] + + +class RandomIdentitySampler(Sampler): + """Randomly samples N identities each with K instances. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + num_instances (int): number of instances per identity in a batch. + """ + + def __init__(self, data_source, batch_size, num_instances): + if batch_size < num_instances: + raise ValueError( + 'batch_size={} must be no less ' + 'than num_instances={}'.format(batch_size, num_instances) + ) + + self.data_source = data_source + self.batch_size = batch_size + self.num_instances = num_instances + self.num_pids_per_batch = self.batch_size // self.num_instances + self.index_dic = defaultdict(list) + for index, items in enumerate(data_source): + pid = items[1] + self.index_dic[pid].append(index) + self.pids = list(self.index_dic.keys()) + assert len(self.pids) >= self.num_pids_per_batch + + # estimate number of examples in an epoch + # TODO: improve precision + self.length = 0 + for pid in self.pids: + idxs = self.index_dic[pid] + num = len(idxs) + if num < self.num_instances: + num = self.num_instances + self.length += num - num % self.num_instances + + def __iter__(self): + batch_idxs_dict = defaultdict(list) + + for pid in self.pids: + idxs = copy.deepcopy(self.index_dic[pid]) + if len(idxs) < self.num_instances: + idxs = np.random.choice( + idxs, size=self.num_instances, replace=True + ) + random.shuffle(idxs) + batch_idxs = [] + for idx in idxs: + batch_idxs.append(idx) + if len(batch_idxs) == self.num_instances: + batch_idxs_dict[pid].append(batch_idxs) + batch_idxs = [] + + avai_pids = copy.deepcopy(self.pids) + final_idxs = [] + + while len(avai_pids) >= self.num_pids_per_batch: + selected_pids = random.sample(avai_pids, self.num_pids_per_batch) + for pid in selected_pids: + batch_idxs = batch_idxs_dict[pid].pop(0) + final_idxs.extend(batch_idxs) + if len(batch_idxs_dict[pid]) == 0: + avai_pids.remove(pid) + + return iter(final_idxs) + + def __len__(self): + return self.length + + +class RandomDomainSampler(Sampler): + """Random domain sampler. + + We consider each camera as a visual domain. + + How does the sampling work: + 1. Randomly sample N cameras (based on the "camid" label). + 2. From each camera, randomly sample K images. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + n_domain (int): number of cameras to sample in a batch. + """ + + def __init__(self, data_source, batch_size, n_domain): + self.data_source = data_source + + # Keep track of image indices for each domain + self.domain_dict = defaultdict(list) + for i, items in enumerate(data_source): + camid = items[2] + self.domain_dict[camid].append(i) + self.domains = list(self.domain_dict.keys()) + + # Make sure each domain can be assigned an equal number of images + if n_domain is None or n_domain <= 0: + n_domain = len(self.domains) + assert batch_size % n_domain == 0 + self.n_img_per_domain = batch_size // n_domain + + self.batch_size = batch_size + self.n_domain = n_domain + self.length = len(list(self.__iter__())) + + def __iter__(self): + domain_dict = copy.deepcopy(self.domain_dict) + final_idxs = [] + stop_sampling = False + + while not stop_sampling: + selected_domains = random.sample(self.domains, self.n_domain) + + for domain in selected_domains: + idxs = domain_dict[domain] + selected_idxs = random.sample(idxs, self.n_img_per_domain) + final_idxs.extend(selected_idxs) + + for idx in selected_idxs: + domain_dict[domain].remove(idx) + + remaining = len(domain_dict[domain]) + if remaining < self.n_img_per_domain: + stop_sampling = True + + return iter(final_idxs) + + def __len__(self): + return self.length + + +class RandomDatasetSampler(Sampler): + """Random dataset sampler. + + How does the sampling work: + 1. Randomly sample N datasets (based on the "dsetid" label). + 2. From each dataset, randomly sample K images. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid, dsetid). + batch_size (int): batch size. + n_dataset (int): number of datasets to sample in a batch. + """ + + def __init__(self, data_source, batch_size, n_dataset): + self.data_source = data_source + + # Keep track of image indices for each dataset + self.dataset_dict = defaultdict(list) + for i, items in enumerate(data_source): + dsetid = items[3] + self.dataset_dict[dsetid].append(i) + self.datasets = list(self.dataset_dict.keys()) + + # Make sure each dataset can be assigned an equal number of images + if n_dataset is None or n_dataset <= 0: + n_dataset = len(self.datasets) + assert batch_size % n_dataset == 0 + self.n_img_per_dset = batch_size // n_dataset + + self.batch_size = batch_size + self.n_dataset = n_dataset + self.length = len(list(self.__iter__())) + + def __iter__(self): + dataset_dict = copy.deepcopy(self.dataset_dict) + final_idxs = [] + stop_sampling = False + + while not stop_sampling: + selected_datasets = random.sample(self.datasets, self.n_dataset) + + for dset in selected_datasets: + idxs = dataset_dict[dset] + selected_idxs = random.sample(idxs, self.n_img_per_dset) + final_idxs.extend(selected_idxs) + + for idx in selected_idxs: + dataset_dict[dset].remove(idx) + + remaining = len(dataset_dict[dset]) + if remaining < self.n_img_per_dset: + stop_sampling = True + + return iter(final_idxs) + + def __len__(self): + return self.length + + +def build_train_sampler( + data_source, + train_sampler, + batch_size=32, + num_instances=4, + num_cams=1, + num_datasets=1, + **kwargs +): + """Builds a training sampler. + + Args: + data_source (list): contains tuples of (img_path(s), pid, camid). + train_sampler (str): sampler name (default: ``RandomSampler``). + batch_size (int, optional): batch size. Default is 32. + num_instances (int, optional): number of instances per identity in a + batch (when using ``RandomIdentitySampler``). Default is 4. + num_cams (int, optional): number of cameras to sample in a batch (when using + ``RandomDomainSampler``). Default is 1. + num_datasets (int, optional): number of datasets to sample in a batch (when + using ``RandomDatasetSampler``). Default is 1. + """ + assert train_sampler in AVAI_SAMPLERS, \ + 'train_sampler must be one of {}, but got {}'.format(AVAI_SAMPLERS, train_sampler) + + if train_sampler == 'RandomIdentitySampler': + sampler = RandomIdentitySampler(data_source, batch_size, num_instances) + + elif train_sampler == 'RandomDomainSampler': + sampler = RandomDomainSampler(data_source, batch_size, num_cams) + + elif train_sampler == 'RandomDatasetSampler': + sampler = RandomDatasetSampler(data_source, batch_size, num_datasets) + + elif train_sampler == 'SequentialSampler': + sampler = SequentialSampler(data_source) + + elif train_sampler == 'RandomSampler': + sampler = RandomSampler(data_source) + + return sampler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py new file mode 100644 index 0000000000..3108b81565 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/data/transforms.py @@ -0,0 +1,373 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import math +import random +from collections import deque +import torch +from PIL import Image +from torchvision.transforms import ( + Resize, Compose, ToTensor, Normalize, ColorJitter, RandomHorizontalFlip +) + + +class Random2DTranslation(object): + """Randomly translates the input image with a probability. + + Specifically, given a predefined shape (height, width), the input is first + resized with a factor of 1.125, leading to (height*1.125, width*1.125), then + a random crop is performed. Such operation is done with a probability. + + Args: + height (int): target image height. + width (int): target image width. + p (float, optional): probability that this operation takes place. + Default is 0.5. + interpolation (int, optional): desired interpolation. Default is + ``PIL.Image.BILINEAR`` + """ + + def __init__(self, height, width, p=0.5, interpolation=Image.BILINEAR): + self.height = height + self.width = width + self.p = p + self.interpolation = interpolation + + def __call__(self, img): + if random.uniform(0, 1) > self.p: + return img.resize((self.width, self.height), self.interpolation) + + new_width, new_height = int(round(self.width * 1.125) + ), int(round(self.height * 1.125)) + resized_img = img.resize((new_width, new_height), self.interpolation) + x_maxrange = new_width - self.width + y_maxrange = new_height - self.height + x1 = int(round(random.uniform(0, x_maxrange))) + y1 = int(round(random.uniform(0, y_maxrange))) + croped_img = resized_img.crop( + (x1, y1, x1 + self.width, y1 + self.height) + ) + return croped_img + + +class RandomErasing(object): + """Randomly erases an image patch. + + Origin: ``_ + + Reference: + Zhong et al. Random Erasing Data Augmentation. + + Args: + probability (float, optional): probability that this operation takes place. + Default is 0.5. + sl (float, optional): min erasing area. + sh (float, optional): max erasing area. + r1 (float, optional): min aspect ratio. + mean (list, optional): erasing value. + """ + + def __init__( + self, + probability=0.5, + sl=0.02, + sh=0.4, + r1=0.3, + mean=[0.4914, 0.4822, 0.4465] + ): + self.probability = probability + self.mean = mean + self.sl = sl + self.sh = sh + self.r1 = r1 + + def __call__(self, img): + if random.uniform(0, 1) > self.probability: + return img + + for attempt in range(100): + area = img.size()[1] * img.size()[2] + + target_area = random.uniform(self.sl, self.sh) * area + aspect_ratio = random.uniform(self.r1, 1 / self.r1) + + h = int(round(math.sqrt(target_area * aspect_ratio))) + w = int(round(math.sqrt(target_area / aspect_ratio))) + + if w < img.size()[2] and h < img.size()[1]: + x1 = random.randint(0, img.size()[1] - h) + y1 = random.randint(0, img.size()[2] - w) + if img.size()[0] == 3: + img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] + img[1, x1:x1 + h, y1:y1 + w] = self.mean[1] + img[2, x1:x1 + h, y1:y1 + w] = self.mean[2] + else: + img[0, x1:x1 + h, y1:y1 + w] = self.mean[0] + return img + + return img + + +class ColorAugmentation(object): + """Randomly alters the intensities of RGB channels. + + Reference: + Krizhevsky et al. ImageNet Classification with Deep ConvolutionalNeural + Networks. NIPS 2012. + + Args: + p (float, optional): probability that this operation takes place. + Default is 0.5. + """ + + def __init__(self, p=0.5): + self.p = p + self.eig_vec = torch.Tensor( + [ + [0.4009, 0.7192, -0.5675], + [-0.8140, -0.0045, -0.5808], + [0.4203, -0.6948, -0.5836], + ] + ) + self.eig_val = torch.Tensor([[0.2175, 0.0188, 0.0045]]) + + def _check_input(self, tensor): + assert tensor.dim() == 3 and tensor.size(0) == 3 + + def __call__(self, tensor): + if random.uniform(0, 1) > self.p: + return tensor + alpha = torch.normal(mean=torch.zeros_like(self.eig_val)) * 0.1 + quatity = torch.mm(self.eig_val * alpha, self.eig_vec) + tensor = tensor + quatity.view(3, 1, 1) + return tensor + + +class RandomPatch(object): + """Random patch data augmentation. + + There is a patch pool that stores randomly extracted pathces from person images. + + For each input image, RandomPatch + 1) extracts a random patch and stores the patch in the patch pool; + 2) randomly selects a patch from the patch pool and pastes it on the + input (at random position) to simulate occlusion. + + Reference: + - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. + - Zhou et al. Learning Generalisable Omni-Scale Representations + for Person Re-Identification. TPAMI, 2021. + """ + + def __init__( + self, + prob_happen=0.5, + pool_capacity=50000, + min_sample_size=100, + patch_min_area=0.01, + patch_max_area=0.5, + patch_min_ratio=0.1, + prob_rotate=0.5, + prob_flip_leftright=0.5, + ): + self.prob_happen = prob_happen + + self.patch_min_area = patch_min_area + self.patch_max_area = patch_max_area + self.patch_min_ratio = patch_min_ratio + + self.prob_rotate = prob_rotate + self.prob_flip_leftright = prob_flip_leftright + + self.patchpool = deque(maxlen=pool_capacity) + self.min_sample_size = min_sample_size + + def generate_wh(self, W, H): + area = W * H + for attempt in range(100): + target_area = random.uniform( + self.patch_min_area, self.patch_max_area + ) * area + aspect_ratio = random.uniform( + self.patch_min_ratio, 1. / self.patch_min_ratio + ) + h = int(round(math.sqrt(target_area * aspect_ratio))) + w = int(round(math.sqrt(target_area / aspect_ratio))) + if w < W and h < H: + return w, h + return None, None + + def transform_patch(self, patch): + if random.uniform(0, 1) > self.prob_flip_leftright: + patch = patch.transpose(Image.FLIP_LEFT_RIGHT) + if random.uniform(0, 1) > self.prob_rotate: + patch = patch.rotate(random.randint(-10, 10)) + return patch + + def __call__(self, img): + W, H = img.size # original image size + + # collect new patch + w, h = self.generate_wh(W, H) + if w is not None and h is not None: + x1 = random.randint(0, W - w) + y1 = random.randint(0, H - h) + new_patch = img.crop((x1, y1, x1 + w, y1 + h)) + self.patchpool.append(new_patch) + + if len(self.patchpool) < self.min_sample_size: + return img + + if random.uniform(0, 1) > self.prob_happen: + return img + + # paste a randomly selected patch on a random position + patch = random.sample(self.patchpool, 1)[0] + patchW, patchH = patch.size + x1 = random.randint(0, W - patchW) + y1 = random.randint(0, H - patchH) + patch = self.transform_patch(patch) + img.paste(patch, (x1, y1)) + + return img + + +def build_transforms( + height, + width, + transforms='random_flip', + norm_mean=[0.485, 0.456, 0.406], + norm_std=[0.229, 0.224, 0.225], + **kwargs +): + """Builds train and test transform functions. + + Args: + height (int): target image height. + width (int): target image width. + transforms (str or list of str, optional): transformations applied to model training. + Default is 'random_flip'. + norm_mean (list or None, optional): normalization mean values. Default is ImageNet means. + norm_std (list or None, optional): normalization standard deviation values. Default is + ImageNet standard deviation values. + """ + if transforms is None: + transforms = [] + + if isinstance(transforms, str): + transforms = [transforms] + + if not isinstance(transforms, list): + raise ValueError( + 'transforms must be a list of strings, but found to be {}'.format( + type(transforms) + ) + ) + + if len(transforms) > 0: + transforms = [t.lower() for t in transforms] + + if norm_mean is None or norm_std is None: + norm_mean = [0.485, 0.456, 0.406] # imagenet mean + norm_std = [0.229, 0.224, 0.225] # imagenet std + normalize = Normalize(mean=norm_mean, std=norm_std) + + print('Building train transforms ...') + transform_tr = [] + + print('+ resize to {}x{}'.format(height, width)) + transform_tr += [Resize((height, width))] + + if 'random_flip' in transforms: + print('+ random flip') + transform_tr += [RandomHorizontalFlip()] + + if 'random_crop' in transforms: + print( + '+ random crop (enlarge to {}x{} and ' + 'crop {}x{})'.format( + int(round(height * 1.125)), int(round(width * 1.125)), height, + width + ) + ) + transform_tr += [Random2DTranslation(height, width)] + + if 'random_patch' in transforms: + print('+ random patch') + transform_tr += [RandomPatch()] + + if 'color_jitter' in transforms: + print('+ color jitter') + transform_tr += [ + ColorJitter(brightness=0.2, contrast=0.15, saturation=0, hue=0) + ] + + print('+ to torch tensor of range [0, 1]') + transform_tr += [ToTensor()] + + print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) + transform_tr += [normalize] + + if 'random_erase' in transforms: + print('+ random erase') + transform_tr += [RandomErasing(mean=norm_mean)] + + transform_tr = Compose(transform_tr) + + print('Building test transforms ...') + print('+ resize to {}x{}'.format(height, width)) + print('+ to torch tensor of range [0, 1]') + print('+ normalization (mean={}, std={})'.format(norm_mean, norm_std)) + + transform_te = Compose([ + Resize((height, width)), + ToTensor(), + normalize, + ]) + + return transform_tr, transform_te diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py new file mode 100644 index 0000000000..7eca9586f7 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/__init__.py @@ -0,0 +1,52 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import + +from .image import ImageSoftmaxEngine, ImageTripletEngine +from .video import VideoSoftmaxEngine, VideoTripletEngine +from .engine import Engine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py new file mode 100644 index 0000000000..4dd250fa95 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/engine.py @@ -0,0 +1,547 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import time +import numpy as np +import os.path as osp +import datetime +from collections import OrderedDict +import torch +from torch.nn import functional as F +from torch.utils.tensorboard import SummaryWriter + +from torchreid import metrics +from torchreid.utils import ( + MetricMeter, AverageMeter, re_ranking, open_all_layers, save_checkpoint, + open_specified_layers, visualize_ranked_results +) +from torchreid.losses import DeepSupervision +import os + + +class Engine(object): + r"""A generic base Engine class for both image- and video-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + use_gpu (bool, optional): use gpu. Default is True. + """ + + def __init__(self, datamanager, use_gpu=False, use_npu=False): + self.datamanager = datamanager + self.train_loader = self.datamanager.train_loader + self.test_loader = self.datamanager.test_loader + self.use_gpu = use_gpu + self.use_npu = use_npu + self.writer = None + self.epoch = 0 + + self.model = None + self.optimizer = None + self.scheduler = None + + self._models = OrderedDict() + self._optims = OrderedDict() + self._scheds = OrderedDict() + + def register_model(self, name='model', model=None, optim=None, sched=None): + if self.__dict__.get('_models') is None: + raise AttributeError( + 'Cannot assign model before super().__init__() call' + ) + + if self.__dict__.get('_optims') is None: + raise AttributeError( + 'Cannot assign optim before super().__init__() call' + ) + + if self.__dict__.get('_scheds') is None: + raise AttributeError( + 'Cannot assign sched before super().__init__() call' + ) + + self._models[name] = model + self._optims[name] = optim + self._scheds[name] = sched + + def get_model_names(self, names=None): + names_real = list(self._models.keys()) + if names is not None: + if not isinstance(names, list): + names = [names] + for name in names: + assert name in names_real + return names + else: + return names_real + + def save_model(self, epoch, rank1, save_dir, is_best=False): + names = self.get_model_names() + + for name in names: + save_checkpoint( + { + 'state_dict': self._models[name].state_dict(), + 'epoch': epoch + 1, + 'rank1': rank1, + 'optimizer': self._optims[name].state_dict(), + 'scheduler': self._scheds[name].state_dict() + }, + osp.join(save_dir, name), + is_best=is_best + ) + + def set_model_mode(self, mode='train', names=None): + assert mode in ['train', 'eval', 'test'] + names = self.get_model_names(names) + + for name in names: + if mode == 'train': + self._models[name].train() + else: + self._models[name].eval() + + def get_current_lr(self, names=None): + names = self.get_model_names(names) + name = names[0] + return self._optims[name].param_groups[-1]['lr'] + + def update_lr(self, names=None): + names = self.get_model_names(names) + + for name in names: + if self._scheds[name] is not None: + self._scheds[name].step() + + def run( + self, + save_dir='log', + max_epoch=0, + start_epoch=0, + print_freq=10, + fixbase_epoch=0, + open_layers=None, + start_eval=0, + eval_freq=-1, + test_only=False, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + r"""A unified pipeline for training and evaluating a model. + + Args: + save_dir (str): directory to save model. + max_epoch (int): maximum epoch. + start_epoch (int, optional): starting epoch. Default is 0. + print_freq (int, optional): print_frequency. Default is 10. + fixbase_epoch (int, optional): number of epochs to train ``open_layers`` (new layers) + while keeping base layers frozen. Default is 0. ``fixbase_epoch`` is counted + in ``max_epoch``. + open_layers (str or list, optional): layers (attribute names) open for training. + start_eval (int, optional): from which epoch to start evaluation. Default is 0. + eval_freq (int, optional): evaluation frequency. Default is -1 (meaning evaluation + is only performed at the end of training). + test_only (bool, optional): if True, only runs evaluation on test datasets. + Default is False. + dist_metric (str, optional): distance metric used to compute distance matrix + between query and gallery. Default is "euclidean". + normalize_feature (bool, optional): performs L2 normalization on feature vectors before + computing feature distance. Default is False. + visrank (bool, optional): visualizes ranked results. Default is False. It is recommended to + enable ``visrank`` when ``test_only`` is True. The ranked images will be saved to + "save_dir/visrank_dataset", e.g. "save_dir/visrank_market1501". + visrank_topk (int, optional): top-k ranked images to be visualized. Default is 10. + use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. + Default is False. This should be enabled when using cuhk03 classic split. + ranks (list, optional): cmc ranks to be computed. Default is [1, 5, 10, 20]. + rerank (bool, optional): uses person re-ranking (by Zhong et al. CVPR'17). + Default is False. This is only enabled when test_only=True. + """ + + if visrank and not test_only: + raise ValueError( + 'visrank can be set to True only if test_only=True' + ) + + if test_only: + self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks, + rerank=rerank + ) + return + + if self.writer is None: + self.writer = SummaryWriter(log_dir=save_dir) + + time_start = time.time() + self.start_epoch = start_epoch + self.max_epoch = max_epoch + print('=> Start training') + + device_num = int(os.environ['device_num']) + device_num = 1 if device_num == -1 else device_num + batch_size = int(os.environ['batch_size']) + total_avg = 0.0 + + for self.epoch in range(self.start_epoch, self.max_epoch): + if os.environ['device_num'] != '-1' and os.environ['device_num'] != '1': + self.datamanager.train_sampler.set_epoch(self.epoch) + eve_time = self.train( + print_freq=print_freq, + fixbase_epoch=fixbase_epoch, + open_layers=open_layers + ) + total_avg += eve_time + print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / eve_time, eve_time)) + + if (self.epoch + 1) >= start_eval \ + and eval_freq > 0 \ + and (self.epoch+1) % eval_freq == 0 \ + and (self.epoch + 1) != self.max_epoch: + rank1 = self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks + ) + self.save_model(self.epoch, rank1, save_dir) + + avg_time = total_avg / (self.max_epoch - self.start_epoch) + + if self.max_epoch > 1: + print('=> Final test') + rank1 = self.test( + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks + ) + self.save_model(self.epoch, rank1, save_dir) + + elapsed = round(time.time() - time_start) + elapsed = str(datetime.timedelta(seconds=elapsed)) + print('Elapsed {}'.format(elapsed)) + + print('FPS@all {:.3f}, TIME@all {:.3f}'.format(device_num * batch_size / avg_time, avg_time)) + + if self.writer is not None: + self.writer.close() + + def train(self, print_freq=10, fixbase_epoch=0, open_layers=None): + losses = MetricMeter() + batch_time = AverageMeter() + data_time = AverageMeter() + + self.set_model_mode('train') + + self.two_stepped_transfer_learning( + self.epoch, fixbase_epoch, open_layers + ) + + self.num_batches = len(self.train_loader) + end = time.time() + for self.batch_idx, data in enumerate(self.train_loader): + data_time.update(time.time() - end) + loss_summary = self.forward_backward(data) + batch_time.update(time.time() - end) + losses.update(loss_summary) + + if (self.batch_idx + 1) % print_freq == 0: + nb_this_epoch = self.num_batches - (self.batch_idx + 1) + nb_future_epochs = ( + self.max_epoch - (self.epoch + 1) + ) * self.num_batches + eta_seconds = batch_time.avg * (nb_this_epoch+nb_future_epochs) + eta_str = str(datetime.timedelta(seconds=int(eta_seconds))) + print( + 'epoch: [{0}/{1}][{2}/{3}]\t' + 'time {batch_time.val:.3f} ({batch_time.avg:.3f})\t' + 'data {data_time.val:.3f} ({data_time.avg:.3f})\t' + 'eta {eta}\t' + '{losses}\t' + 'lr {lr:.6f}'.format( + self.epoch + 1, + self.max_epoch, + self.batch_idx + 1, + self.num_batches, + batch_time=batch_time, + data_time=data_time, + eta=eta_str, + losses=losses, + lr=self.get_current_lr() + ) + ) + + if self.writer is not None: + n_iter = self.epoch * self.num_batches + self.batch_idx + self.writer.add_scalar('Train/time', batch_time.avg, n_iter) + self.writer.add_scalar('Train/data', data_time.avg, n_iter) + for name, meter in losses.meters.items(): + self.writer.add_scalar('Train/' + name, meter.avg, n_iter) + self.writer.add_scalar( + 'Train/lr', self.get_current_lr(), n_iter + ) + + end = time.time() + + if self.batch_idx == 4: # 前5个step不记录时间 + batch_time.reset() + + self.update_lr() + return batch_time.avg + + def forward_backward(self, data): + raise NotImplementedError + + def test( + self, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + save_dir='', + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + r"""Tests model on target datasets. + + .. note:: + + This function has been called in ``run()``. + + .. note:: + + The test pipeline implemented in this function suits both image- and + video-reid. In general, a subclass of Engine only needs to re-implement + ``extract_features()`` and ``parse_data_for_eval()`` (most of the time), + but not a must. Please refer to the source code for more details. + """ + self.set_model_mode('eval') + targets = list(self.test_loader.keys()) + + for name in targets: + domain = 'source' if name in self.datamanager.sources else 'target' + print('##### Evaluating {} ({}) #####'.format(name, domain)) + query_loader = self.test_loader[name]['query'] + gallery_loader = self.test_loader[name]['gallery'] + rank1, mAP = self._evaluate( + dataset_name=name, + query_loader=query_loader, + gallery_loader=gallery_loader, + dist_metric=dist_metric, + normalize_feature=normalize_feature, + visrank=visrank, + visrank_topk=visrank_topk, + save_dir=save_dir, + use_metric_cuhk03=use_metric_cuhk03, + ranks=ranks, + rerank=rerank + ) + + if self.writer is not None: + self.writer.add_scalar(f'Test/{name}/rank1', rank1, self.epoch) + self.writer.add_scalar(f'Test/{name}/mAP', mAP, self.epoch) + + return rank1 + + @torch.no_grad() + def _evaluate( + self, + dataset_name='', + query_loader=None, + gallery_loader=None, + dist_metric='euclidean', + normalize_feature=False, + visrank=False, + visrank_topk=10, + save_dir='', + use_metric_cuhk03=False, + ranks=[1, 5, 10, 20], + rerank=False + ): + batch_time = AverageMeter() + + def _feature_extraction(data_loader): + f_, pids_, camids_ = [], [], [] + for batch_idx, data in enumerate(data_loader): + imgs, pids, camids = self.parse_data_for_eval(data) + if self.use_gpu: + imgs = imgs.cuda() + elif self.use_npu: + imgs = imgs.npu() + end = time.time() + features = self.extract_features(imgs) + batch_time.update(time.time() - end) + features = features.cpu().clone() + f_.append(features) + pids_.extend(pids) + camids_.extend(camids) + f_ = torch.cat(f_, 0) + pids_ = np.asarray(pids_) + camids_ = np.asarray(camids_) + return f_, pids_, camids_ + + print('Extracting features from query set ...') + qf, q_pids, q_camids = _feature_extraction(query_loader) + print('Done, obtained {}-by-{} matrix'.format(qf.size(0), qf.size(1))) + + print('Extracting features from gallery set ...') + gf, g_pids, g_camids = _feature_extraction(gallery_loader) + print('Done, obtained {}-by-{} matrix'.format(gf.size(0), gf.size(1))) + + print('Speed: {:.4f} sec/batch'.format(batch_time.avg)) + + if normalize_feature: + print('Normalzing features with L2 norm ...') + qf = F.normalize(qf, p=2, dim=1) + gf = F.normalize(gf, p=2, dim=1) + + print( + 'Computing distance matrix with metric={} ...'.format(dist_metric) + ) + distmat = metrics.compute_distance_matrix(qf, gf, dist_metric) + distmat = distmat.numpy() + + if rerank: + print('Applying person re-ranking ...') + distmat_qq = metrics.compute_distance_matrix(qf, qf, dist_metric) + distmat_gg = metrics.compute_distance_matrix(gf, gf, dist_metric) + distmat = re_ranking(distmat, distmat_qq, distmat_gg) + + print('Computing CMC and mAP ...') + cmc, mAP = metrics.evaluate_rank( + distmat, + q_pids, + g_pids, + q_camids, + g_camids, + use_metric_cuhk03=use_metric_cuhk03 + ) + + print('** Results **') + print('mAP: {:.1%}'.format(mAP)) + print('CMC curve') + for r in ranks: + print('Rank-{:<3}: {:.1%}'.format(r, cmc[r - 1])) + + if visrank: + visualize_ranked_results( + distmat, + self.datamanager.fetch_test_loaders(dataset_name), + self.datamanager.data_type, + width=self.datamanager.width, + height=self.datamanager.height, + save_dir=osp.join(save_dir, 'visrank_' + dataset_name), + topk=visrank_topk + ) + + return cmc[0], mAP + + def compute_loss(self, criterion, outputs, targets): + if isinstance(outputs, (tuple, list)): + loss = DeepSupervision(criterion, outputs, targets) + else: + loss = criterion(outputs, targets) + return loss + + def extract_features(self, input): + return self.model(input) + + def parse_data_for_train(self, data): + imgs = data['img'] + pids = data['pid'] + return imgs, pids + + def parse_data_for_eval(self, data): + imgs = data['img'] + pids = data['pid'] + camids = data['camid'] + return imgs, pids, camids + + def two_stepped_transfer_learning( + self, epoch, fixbase_epoch, open_layers, model=None + ): + """Two-stepped transfer learning. + + The idea is to freeze base layers for a certain number of epochs + and then open all layers for training. + + Reference: https://arxiv.org/abs/1611.05244 + """ + model = self.model if model is None else model + if model is None: + return + + if (epoch + 1) <= fixbase_epoch and open_layers is not None: + print( + '* Only train {} (epoch: {}/{})'.format( + open_layers, epoch + 1, fixbase_epoch + ) + ) + open_specified_layers(model, open_layers) + else: + open_all_layers(model) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py new file mode 100644 index 0000000000..fef3e0a2e3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/__init__.py @@ -0,0 +1,51 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .softmax import ImageSoftmaxEngine +from .triplet import ImageTripletEngine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py new file mode 100644 index 0000000000..0a836672d5 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/softmax.py @@ -0,0 +1,157 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + +from torchreid import metrics +from torchreid.losses import CrossEntropyLoss + +from ..engine import Engine + +from apex import amp + +class ImageSoftmaxEngine(Engine): + r"""Softmax-loss engine for image-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + model (nn.Module): model instance. + optimizer (Optimizer): an Optimizer. + scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. + use_gpu (bool, optional): use gpu. Default is True. + label_smooth (bool, optional): use label smoothing regularizer. Default is True. + + Examples:: + + import torchreid + datamanager = torchreid.data.ImageDataManager( + root='path/to/reid-data', + sources='market1501', + height=256, + width=128, + combineall=False, + batch_size=32 + ) + model = torchreid.models.build_model( + name='resnet50', + num_classes=datamanager.num_train_pids, + loss='softmax' + ) + model = model.cuda() + optimizer = torchreid.optim.build_optimizer( + model, optim='adam', lr=0.0003 + ) + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler='single_step', + stepsize=20 + ) + engine = torchreid.engine.ImageSoftmaxEngine( + datamanager, model, optimizer, scheduler=scheduler + ) + engine.run( + max_epoch=60, + save_dir='log/resnet50-softmax-market1501', + print_freq=10 + ) + """ + + def __init__( + self, + datamanager, + model, + optimizer, + scheduler=None, + use_gpu=False, + use_npu=False, + label_smooth=True, + use_amp=False + ): + super(ImageSoftmaxEngine, self).__init__(datamanager, use_gpu, use_npu) + + self.model = model + self.optimizer = optimizer + self.scheduler = scheduler + self.register_model('model', model, optimizer, scheduler) + self.use_amp = use_amp + + + self.criterion = CrossEntropyLoss( + num_classes=self.datamanager.num_train_pids, + use_gpu=self.use_gpu, + use_npu=self.use_npu, + label_smooth=label_smooth + ) + + def forward_backward(self, data): + imgs, pids = self.parse_data_for_train(data) + + if self.use_gpu: + imgs = imgs.cuda() + pids = pids.cuda() + elif self.use_npu: + imgs = imgs.npu() + pids = pids.npu() + + outputs = self.model(imgs) + loss = self.compute_loss(self.criterion, outputs, pids) + + self.optimizer.zero_grad() + if self.use_amp: + with amp.scale_loss(loss, self.optimizer) as scaled_loss: + scaled_loss.backward() + else: + loss.backward() + self.optimizer.step() + + loss_summary = { + 'loss': loss.item(), + 'acc': metrics.accuracy(outputs, pids)[0].item() + } + + return loss_summary diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py new file mode 100644 index 0000000000..64a807734b --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/image/triplet.py @@ -0,0 +1,169 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + +from torchreid import metrics +from torchreid.losses import TripletLoss, CrossEntropyLoss + +from ..engine import Engine + + +class ImageTripletEngine(Engine): + r"""Triplet-loss engine for image-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + model (nn.Module): model instance. + optimizer (Optimizer): an Optimizer. + margin (float, optional): margin for triplet loss. Default is 0.3. + weight_t (float, optional): weight for triplet loss. Default is 1. + weight_x (float, optional): weight for softmax loss. Default is 1. + scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. + use_gpu (bool, optional): use gpu. Default is True. + label_smooth (bool, optional): use label smoothing regularizer. Default is True. + + Examples:: + + import torchreid + datamanager = torchreid.data.ImageDataManager( + root='path/to/reid-data', + sources='market1501', + height=256, + width=128, + combineall=False, + batch_size=32, + num_instances=4, + train_sampler='RandomIdentitySampler' # this is important + ) + model = torchreid.models.build_model( + name='resnet50', + num_classes=datamanager.num_train_pids, + loss='triplet' + ) + model = model.cuda() + optimizer = torchreid.optim.build_optimizer( + model, optim='adam', lr=0.0003 + ) + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler='single_step', + stepsize=20 + ) + engine = torchreid.engine.ImageTripletEngine( + datamanager, model, optimizer, margin=0.3, + weight_t=0.7, weight_x=1, scheduler=scheduler + ) + engine.run( + max_epoch=60, + save_dir='log/resnet50-triplet-market1501', + print_freq=10 + ) + """ + + def __init__( + self, + datamanager, + model, + optimizer, + margin=0.3, + weight_t=1, + weight_x=1, + scheduler=None, + use_gpu=True, + label_smooth=True + ): + super(ImageTripletEngine, self).__init__(datamanager, use_gpu) + + self.model = model + self.optimizer = optimizer + self.scheduler = scheduler + self.register_model('model', model, optimizer, scheduler) + + assert weight_t >= 0 and weight_x >= 0 + assert weight_t + weight_x > 0 + self.weight_t = weight_t + self.weight_x = weight_x + + self.criterion_t = TripletLoss(margin=margin) + self.criterion_x = CrossEntropyLoss( + num_classes=self.datamanager.num_train_pids, + use_gpu=self.use_gpu, + label_smooth=label_smooth + ) + + def forward_backward(self, data): + imgs, pids = self.parse_data_for_train(data) + + if self.use_gpu: + imgs = imgs.cuda() + pids = pids.cuda() + + outputs, features = self.model(imgs) + + loss = 0 + loss_summary = {} + + if self.weight_t > 0: + loss_t = self.compute_loss(self.criterion_t, features, pids) + loss += self.weight_t * loss_t + loss_summary['loss_t'] = loss_t.item() + + if self.weight_x > 0: + loss_x = self.compute_loss(self.criterion_x, outputs, pids) + loss += self.weight_x * loss_x + loss_summary['loss_x'] = loss_x.item() + loss_summary['acc'] = metrics.accuracy(outputs, pids)[0].item() + + assert loss_summary + + self.optimizer.zero_grad() + loss.backward() + self.optimizer.step() + + return loss_summary diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py new file mode 100644 index 0000000000..28fc05991d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/__init__.py @@ -0,0 +1,51 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .softmax import VideoSoftmaxEngine +from .triplet import VideoTripletEngine diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py new file mode 100644 index 0000000000..44c42bb795 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/softmax.py @@ -0,0 +1,156 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch + +from torchreid.engine.image import ImageSoftmaxEngine + + +class VideoSoftmaxEngine(ImageSoftmaxEngine): + """Softmax-loss engine for video-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + model (nn.Module): model instance. + optimizer (Optimizer): an Optimizer. + scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. + use_gpu (bool, optional): use gpu. Default is True. + label_smooth (bool, optional): use label smoothing regularizer. Default is True. + pooling_method (str, optional): how to pool features for a tracklet. + Default is "avg" (average). Choices are ["avg", "max"]. + + Examples:: + + import torch + import torchreid + # Each batch contains batch_size*seq_len images + datamanager = torchreid.data.VideoDataManager( + root='path/to/reid-data', + sources='mars', + height=256, + width=128, + combineall=False, + batch_size=8, # number of tracklets + seq_len=15 # number of images in each tracklet + ) + model = torchreid.models.build_model( + name='resnet50', + num_classes=datamanager.num_train_pids, + loss='softmax' + ) + model = model.cuda() + optimizer = torchreid.optim.build_optimizer( + model, optim='adam', lr=0.0003 + ) + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler='single_step', + stepsize=20 + ) + engine = torchreid.engine.VideoSoftmaxEngine( + datamanager, model, optimizer, scheduler=scheduler, + pooling_method='avg' + ) + engine.run( + max_epoch=60, + save_dir='log/resnet50-softmax-mars', + print_freq=10 + ) + """ + + def __init__( + self, + datamanager, + model, + optimizer, + scheduler=None, + use_gpu=True, + label_smooth=True, + pooling_method='avg' + ): + super(VideoSoftmaxEngine, self).__init__( + datamanager, + model, + optimizer, + scheduler=scheduler, + use_gpu=use_gpu, + label_smooth=label_smooth + ) + self.pooling_method = pooling_method + + def parse_data_for_train(self, data): + imgs = data['img'] + pids = data['pid'] + if imgs.dim() == 5: + # b: batch size + # s: sqeuence length + # c: channel depth + # h: height + # w: width + b, s, c, h, w = imgs.size() + imgs = imgs.view(b * s, c, h, w) + pids = pids.view(b, 1).expand(b, s) + pids = pids.contiguous().view(b * s) + return imgs, pids + + def extract_features(self, input): + # b: batch size + # s: sqeuence length + # c: channel depth + # h: height + # w: width + b, s, c, h, w = input.size() + input = input.view(b * s, c, h, w) + features = self.model(input) + features = features.view(b, s, -1) + if self.pooling_method == 'avg': + features = torch.mean(features, 1) + else: + features = torch.max(features, 1)[0] + return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py new file mode 100644 index 0000000000..5604715d4a --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/engine/video/triplet.py @@ -0,0 +1,169 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch + +from torchreid.engine.image import ImageTripletEngine + + +class VideoTripletEngine(ImageTripletEngine): + """Triplet-loss engine for video-reid. + + Args: + datamanager (DataManager): an instance of ``torchreid.data.ImageDataManager`` + or ``torchreid.data.VideoDataManager``. + model (nn.Module): model instance. + optimizer (Optimizer): an Optimizer. + margin (float, optional): margin for triplet loss. Default is 0.3. + weight_t (float, optional): weight for triplet loss. Default is 1. + weight_x (float, optional): weight for softmax loss. Default is 1. + scheduler (LRScheduler, optional): if None, no learning rate decay will be performed. + use_gpu (bool, optional): use gpu. Default is True. + label_smooth (bool, optional): use label smoothing regularizer. Default is True. + pooling_method (str, optional): how to pool features for a tracklet. + Default is "avg" (average). Choices are ["avg", "max"]. + + Examples:: + + import torch + import torchreid + # Each batch contains batch_size*seq_len images + # Each identity is sampled with num_instances tracklets + datamanager = torchreid.data.VideoDataManager( + root='path/to/reid-data', + sources='mars', + height=256, + width=128, + combineall=False, + num_instances=4, + train_sampler='RandomIdentitySampler' + batch_size=8, # number of tracklets + seq_len=15 # number of images in each tracklet + ) + model = torchreid.models.build_model( + name='resnet50', + num_classes=datamanager.num_train_pids, + loss='triplet' + ) + model = model.cuda() + optimizer = torchreid.optim.build_optimizer( + model, optim='adam', lr=0.0003 + ) + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler='single_step', + stepsize=20 + ) + engine = torchreid.engine.VideoTripletEngine( + datamanager, model, optimizer, margin=0.3, + weight_t=0.7, weight_x=1, scheduler=scheduler, + pooling_method='avg' + ) + engine.run( + max_epoch=60, + save_dir='log/resnet50-triplet-mars', + print_freq=10 + ) + """ + + def __init__( + self, + datamanager, + model, + optimizer, + margin=0.3, + weight_t=1, + weight_x=1, + scheduler=None, + use_gpu=True, + label_smooth=True, + pooling_method='avg' + ): + super(VideoTripletEngine, self).__init__( + datamanager, + model, + optimizer, + margin=margin, + weight_t=weight_t, + weight_x=weight_x, + scheduler=scheduler, + use_gpu=use_gpu, + label_smooth=label_smooth + ) + self.pooling_method = pooling_method + + def parse_data_for_train(self, data): + imgs = data['img'] + pids = data['pid'] + if imgs.dim() == 5: + # b: batch size + # s: sqeuence length + # c: channel depth + # h: height + # w: width + b, s, c, h, w = imgs.size() + imgs = imgs.view(b * s, c, h, w) + pids = pids.view(b, 1).expand(b, s) + pids = pids.contiguous().view(b * s) + return imgs, pids + + def extract_features(self, input): + # b: batch size + # s: sqeuence length + # c: channel depth + # h: height + # w: width + b, s, c, h, w = input.size() + input = input.view(b * s, c, h, w) + features = self.model(input) + features = features.view(b, s, -1) + if self.pooling_method == 'avg': + features = torch.mean(features, 1) + else: + features = torch.max(features, 1)[0] + return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py new file mode 100644 index 0000000000..0376bc80de --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/__init__.py @@ -0,0 +1,68 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + +from .cross_entropy_loss import CrossEntropyLoss +from .hard_mine_triplet_loss import TripletLoss + + +def DeepSupervision(criterion, xs, y): + """DeepSupervision + + Applies criterion to each element in a list. + + Args: + criterion: loss function + xs: tuple of inputs + y: ground truth + """ + loss = 0. + for x in xs: + loss += criterion(x, y) + loss /= len(xs) + return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py new file mode 100644 index 0000000000..d043691331 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/cross_entropy_loss.py @@ -0,0 +1,100 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn + + +class CrossEntropyLoss(nn.Module): + r"""Cross entropy loss with label smoothing regularizer. + + Reference: + Szegedy et al. Rethinking the Inception Architecture for Computer Vision. CVPR 2016. + + With label smoothing, the label :math:`y` for a class is computed by + + .. math:: + \begin{equation} + (1 - \eps) \times y + \frac{\eps}{K}, + \end{equation} + + where :math:`K` denotes the number of classes and :math:`\eps` is a weight. When + :math:`\eps = 0`, the loss function reduces to the normal cross entropy. + + Args: + num_classes (int): number of classes. + eps (float, optional): weight. Default is 0.1. + use_gpu (bool, optional): whether to use gpu devices. Default is True. + label_smooth (bool, optional): whether to apply label smoothing. Default is True. + """ + + def __init__(self, num_classes, eps=0.1, use_gpu=False, use_npu=False, label_smooth=True): + super(CrossEntropyLoss, self).__init__() + self.num_classes = num_classes + self.eps = eps if label_smooth else 0 + self.use_gpu = use_gpu + self.use_npu = use_npu + self.logsoftmax = nn.LogSoftmax(dim=1) + + def forward(self, inputs, targets): + """ + Args: + inputs (torch.Tensor): prediction matrix (before softmax) with + shape (batch_size, num_classes). + targets (torch.LongTensor): ground truth labels with shape (batch_size). + Each position contains the label index. + """ + log_probs = self.logsoftmax(inputs) + zeros = torch.zeros(log_probs.size()) + targets = zeros.scatter_(1, targets.unsqueeze(1).data.cpu(), 1) + if self.use_gpu: + targets = targets.cuda() + elif self.use_npu: + targets = targets.npu() + targets = (1 - self.eps) * targets + self.eps / self.num_classes + return (-targets * log_probs).mean(0).sum() diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py new file mode 100644 index 0000000000..ac2d927c92 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/losses/hard_mine_triplet_loss.py @@ -0,0 +1,95 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn + + +class TripletLoss(nn.Module): + """Triplet loss with hard positive/negative mining. + + Reference: + Hermans et al. In Defense of the Triplet Loss for Person Re-Identification. arXiv:1703.07737. + + Imported from ``_. + + Args: + margin (float, optional): margin for triplet. Default is 0.3. + """ + + def __init__(self, margin=0.3): + super(TripletLoss, self).__init__() + self.margin = margin + self.ranking_loss = nn.MarginRankingLoss(margin=margin) + + def forward(self, inputs, targets): + """ + Args: + inputs (torch.Tensor): feature matrix with shape (batch_size, feat_dim). + targets (torch.LongTensor): ground truth labels with shape (num_classes). + """ + n = inputs.size(0) + + # Compute pairwise distance, replace by the official when merged + dist = torch.pow(inputs, 2).sum(dim=1, keepdim=True).expand(n, n) + dist = dist + dist.t() + dist.addmm_(inputs, inputs.t(), beta=1, alpha=-2) + dist = dist.clamp(min=1e-12).sqrt() # for numerical stability + + # For each anchor, find the hardest positive and negative + mask = targets.expand(n, n).eq(targets.expand(n, n).t()) + dist_ap, dist_an = [], [] + for i in range(n): + dist_ap.append(dist[i][mask[i]].max().unsqueeze(0)) + dist_an.append(dist[i][mask[i] == 0].min().unsqueeze(0)) + dist_ap = torch.cat(dist_ap) + dist_an = torch.cat(dist_an) + + # Compute ranking hinge loss + y = torch.ones_like(dist_an) + return self.ranking_loss(dist_an, dist_ap, y) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py new file mode 100644 index 0000000000..b1c17830fa --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/__init__.py @@ -0,0 +1,52 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .rank import evaluate_rank +from .accuracy import accuracy +from .distance import compute_distance_matrix diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py new file mode 100644 index 0000000000..c5145818b8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/accuracy.py @@ -0,0 +1,84 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import + + +def accuracy(output, target, topk=(1, )): + """Computes the accuracy over the k top predictions for + the specified values of k. + + Args: + output (torch.Tensor): prediction matrix with shape (batch_size, num_classes). + target (torch.LongTensor): ground truth labels with shape (batch_size). + topk (tuple, optional): accuracy at top-k will be computed. For example, + topk=(1, 5) means accuracy at top-1 and top-5 will be computed. + + Returns: + list: accuracy at top-k. + + Examples:: + >>> from torchreid import metrics + >>> metrics.accuracy(output, target) + """ + maxk = max(topk) + batch_size = target.size(0) + + if isinstance(output, (tuple, list)): + output = output[0] + + _, pred = output.topk(maxk, 1, True, True) + pred = pred.t() + correct = pred.eq(target.view(1, -1).expand_as(pred)) + + res = [] + for k in topk: + correct_k = correct[:k].view(-1).float().sum(0, keepdim=True) + acc = correct_k.mul_(100.0 / batch_size) + res.append(acc) + + return res diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py new file mode 100644 index 0000000000..f209c03fa2 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/distance.py @@ -0,0 +1,127 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import torch +from torch.nn import functional as F + + +def compute_distance_matrix(input1, input2, metric='euclidean'): + """A wrapper function for computing distance matrix. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + metric (str, optional): "euclidean" or "cosine". + Default is "euclidean". + + Returns: + torch.Tensor: distance matrix. + + Examples:: + >>> from torchreid import metrics + >>> input1 = torch.rand(10, 2048) + >>> input2 = torch.rand(100, 2048) + >>> distmat = metrics.compute_distance_matrix(input1, input2) + >>> distmat.size() # (10, 100) + """ + # check input + assert isinstance(input1, torch.Tensor) + assert isinstance(input2, torch.Tensor) + assert input1.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( + input1.dim() + ) + assert input2.dim() == 2, 'Expected 2-D tensor, but got {}-D'.format( + input2.dim() + ) + assert input1.size(1) == input2.size(1) + + if metric == 'euclidean': + distmat = euclidean_squared_distance(input1, input2) + elif metric == 'cosine': + distmat = cosine_distance(input1, input2) + else: + raise ValueError( + 'Unknown distance metric: {}. ' + 'Please choose either "euclidean" or "cosine"'.format(metric) + ) + + return distmat + + +def euclidean_squared_distance(input1, input2): + """Computes euclidean squared distance. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + + Returns: + torch.Tensor: distance matrix. + """ + m, n = input1.size(0), input2.size(0) + mat1 = torch.pow(input1, 2).sum(dim=1, keepdim=True).expand(m, n) + mat2 = torch.pow(input2, 2).sum(dim=1, keepdim=True).expand(n, m).t() + distmat = mat1 + mat2 + distmat.addmm_(input1, input2.t(), beta=1, alpha=-2) + return distmat + + +def cosine_distance(input1, input2): + """Computes cosine distance. + + Args: + input1 (torch.Tensor): 2-D feature matrix. + input2 (torch.Tensor): 2-D feature matrix. + + Returns: + torch.Tensor: distance matrix. + """ + input1_normed = F.normalize(input1, p=2, dim=1) + input2_normed = F.normalize(input2, p=2, dim=1) + distmat = 1 - torch.mm(input1_normed, input2_normed.t()) + return distmat diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py new file mode 100644 index 0000000000..8e9fc70253 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank.py @@ -0,0 +1,254 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import numpy as np +import warnings +from collections import defaultdict + +try: + from torchreid.metrics.rank_cylib.rank_cy import evaluate_cy + IS_CYTHON_AVAI = True +except ImportError: + IS_CYTHON_AVAI = False + warnings.warn( + 'Cython evaluation (very fast so highly recommended) is ' + 'unavailable, now use python evaluation.' + ) + + +def eval_cuhk03(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): + """Evaluation with cuhk03 metric + Key: one image for each gallery identity is randomly sampled for each query identity. + Random sampling is performed num_repeats times. + """ + num_repeats = 10 + num_q, num_g = distmat.shape + + if num_g < max_rank: + max_rank = num_g + print( + 'Note: number of gallery samples is quite small, got {}'. + format(num_g) + ) + + indices = np.argsort(distmat, axis=1) + matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) + + # compute cmc curve for each query + all_cmc = [] + all_AP = [] + num_valid_q = 0. # number of valid query + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + order = indices[q_idx] + remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) + keep = np.invert(remove) + + # compute cmc curve + raw_cmc = matches[q_idx][ + keep] # binary vector, positions with value 1 are correct matches + if not np.any(raw_cmc): + # this condition is true when query identity does not appear in gallery + continue + + kept_g_pids = g_pids[order][keep] + g_pids_dict = defaultdict(list) + for idx, pid in enumerate(kept_g_pids): + g_pids_dict[pid].append(idx) + + cmc = 0. + for repeat_idx in range(num_repeats): + mask = np.zeros(len(raw_cmc), dtype=np.bool) + for _, idxs in g_pids_dict.items(): + # randomly sample one image for each gallery person + rnd_idx = np.random.choice(idxs) + mask[rnd_idx] = True + masked_raw_cmc = raw_cmc[mask] + _cmc = masked_raw_cmc.cumsum() + _cmc[_cmc > 1] = 1 + cmc += _cmc[:max_rank].astype(np.float32) + + cmc /= num_repeats + all_cmc.append(cmc) + # compute AP + num_rel = raw_cmc.sum() + tmp_cmc = raw_cmc.cumsum() + tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] + tmp_cmc = np.asarray(tmp_cmc) * raw_cmc + AP = tmp_cmc.sum() / num_rel + all_AP.append(AP) + num_valid_q += 1. + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + all_cmc = np.asarray(all_cmc).astype(np.float32) + all_cmc = all_cmc.sum(0) / num_valid_q + mAP = np.mean(all_AP) + + return all_cmc, mAP + + +def eval_market1501(distmat, q_pids, g_pids, q_camids, g_camids, max_rank): + """Evaluation with market1501 metric + Key: for each query identity, its gallery images from the same camera view are discarded. + """ + num_q, num_g = distmat.shape + + if num_g < max_rank: + max_rank = num_g + print( + 'Note: number of gallery samples is quite small, got {}'. + format(num_g) + ) + + indices = np.argsort(distmat, axis=1) + matches = (g_pids[indices] == q_pids[:, np.newaxis]).astype(np.int32) + + # compute cmc curve for each query + all_cmc = [] + all_AP = [] + num_valid_q = 0. # number of valid query + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + order = indices[q_idx] + remove = (g_pids[order] == q_pid) & (g_camids[order] == q_camid) + keep = np.invert(remove) + + # compute cmc curve + raw_cmc = matches[q_idx][ + keep] # binary vector, positions with value 1 are correct matches + if not np.any(raw_cmc): + # this condition is true when query identity does not appear in gallery + continue + + cmc = raw_cmc.cumsum() + cmc[cmc > 1] = 1 + + all_cmc.append(cmc[:max_rank]) + num_valid_q += 1. + + # compute average precision + # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision + num_rel = raw_cmc.sum() + tmp_cmc = raw_cmc.cumsum() + tmp_cmc = [x / (i+1.) for i, x in enumerate(tmp_cmc)] + tmp_cmc = np.asarray(tmp_cmc) * raw_cmc + AP = tmp_cmc.sum() / num_rel + all_AP.append(AP) + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + all_cmc = np.asarray(all_cmc).astype(np.float32) + all_cmc = all_cmc.sum(0) / num_valid_q + mAP = np.mean(all_AP) + + return all_cmc, mAP + + +def evaluate_py( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03 +): + if use_metric_cuhk03: + return eval_cuhk03( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank + ) + else: + return eval_market1501( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank + ) + + +def evaluate_rank( + distmat, + q_pids, + g_pids, + q_camids, + g_camids, + max_rank=50, + use_metric_cuhk03=False, + use_cython=True +): + """Evaluates CMC rank. + + Args: + distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). + q_pids (numpy.ndarray): 1-D array containing person identities + of each query instance. + g_pids (numpy.ndarray): 1-D array containing person identities + of each gallery instance. + q_camids (numpy.ndarray): 1-D array containing camera views under + which each query instance is captured. + g_camids (numpy.ndarray): 1-D array containing camera views under + which each gallery instance is captured. + max_rank (int, optional): maximum CMC rank to be computed. Default is 50. + use_metric_cuhk03 (bool, optional): use single-gallery-shot setting for cuhk03. + Default is False. This should be enabled when using cuhk03 classic split. + use_cython (bool, optional): use cython code for evaluation. Default is True. + This is highly recommended as the cython code can speed up the cmc computation + by more than 10x. This requires Cython to be installed. + """ + if use_cython and IS_CYTHON_AVAI: + return evaluate_cy( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, + use_metric_cuhk03 + ) + else: + return evaluate_py( + distmat, q_pids, g_pids, q_camids, g_camids, max_rank, + use_metric_cuhk03 + ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile new file mode 100644 index 0000000000..d49e655f85 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/Makefile @@ -0,0 +1,6 @@ +all: + $(PYTHON) setup.py build_ext --inplace + rm -rf build +clean: + rm -rf build + rm -f rank_cy.c *.so \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py new file mode 100644 index 0000000000..f3bbfaaf4f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/__init__.py @@ -0,0 +1,47 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx new file mode 100644 index 0000000000..b4a8690e57 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/rank_cy.pyx @@ -0,0 +1,251 @@ +# cython: boundscheck=False, wraparound=False, nonecheck=False, cdivision=True + +from __future__ import print_function +import numpy as np +from libc.stdint cimport int64_t, uint64_t + +import cython + +cimport numpy as np + +import random +from collections import defaultdict + +""" +Compiler directives: +https://github.com/cython/cython/wiki/enhancements-compilerdirectives + +Cython tutorial: +https://cython.readthedocs.io/en/latest/src/userguide/numpy_tutorial.html + +Credit to https://github.com/luzai +""" + + +# Main interface +cpdef evaluate_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=False): + distmat = np.asarray(distmat, dtype=np.float32) + q_pids = np.asarray(q_pids, dtype=np.int64) + g_pids = np.asarray(g_pids, dtype=np.int64) + q_camids = np.asarray(q_camids, dtype=np.int64) + g_camids = np.asarray(g_camids, dtype=np.int64) + if use_metric_cuhk03: + return eval_cuhk03_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank) + return eval_market1501_cy(distmat, q_pids, g_pids, q_camids, g_camids, max_rank) + + +cpdef eval_cuhk03_cy(float[:,:] distmat, int64_t[:] q_pids, int64_t[:]g_pids, + int64_t[:]q_camids, int64_t[:]g_camids, int64_t max_rank): + + cdef int64_t num_q = distmat.shape[0] + cdef int64_t num_g = distmat.shape[1] + + if num_g < max_rank: + max_rank = num_g + print('Note: number of gallery samples is quite small, got {}'.format(num_g)) + + cdef: + int64_t num_repeats = 10 + int64_t[:,:] indices = np.argsort(distmat, axis=1) + int64_t[:,:] matches = (np.asarray(g_pids)[np.asarray(indices)] == np.asarray(q_pids)[:, np.newaxis]).astype(np.int64) + + float[:,:] all_cmc = np.zeros((num_q, max_rank), dtype=np.float32) + float[:] all_AP = np.zeros(num_q, dtype=np.float32) + float num_valid_q = 0. # number of valid query + + int64_t q_idx, q_pid, q_camid, g_idx + int64_t[:] order = np.zeros(num_g, dtype=np.int64) + int64_t keep + + float[:] raw_cmc = np.zeros(num_g, dtype=np.float32) # binary vector, positions with value 1 are correct matches + float[:] masked_raw_cmc = np.zeros(num_g, dtype=np.float32) + float[:] cmc, masked_cmc + int64_t num_g_real, num_g_real_masked, rank_idx, rnd_idx + uint64_t meet_condition + float AP + int64_t[:] kept_g_pids, mask + + float num_rel + float[:] tmp_cmc = np.zeros(num_g, dtype=np.float32) + float tmp_cmc_sum + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + for g_idx in range(num_g): + order[g_idx] = indices[q_idx, g_idx] + num_g_real = 0 + meet_condition = 0 + kept_g_pids = np.zeros(num_g, dtype=np.int64) + + for g_idx in range(num_g): + if (g_pids[order[g_idx]] != q_pid) or (g_camids[order[g_idx]] != q_camid): + raw_cmc[num_g_real] = matches[q_idx][g_idx] + kept_g_pids[num_g_real] = g_pids[order[g_idx]] + num_g_real += 1 + if matches[q_idx][g_idx] > 1e-31: + meet_condition = 1 + + if not meet_condition: + # this condition is true when query identity does not appear in gallery + continue + + # cuhk03-specific setting + g_pids_dict = defaultdict(list) # overhead! + for g_idx in range(num_g_real): + g_pids_dict[kept_g_pids[g_idx]].append(g_idx) + + cmc = np.zeros(max_rank, dtype=np.float32) + for _ in range(num_repeats): + mask = np.zeros(num_g_real, dtype=np.int64) + + for _, idxs in g_pids_dict.items(): + # randomly sample one image for each gallery person + rnd_idx = np.random.choice(idxs) + #rnd_idx = idxs[0] # use deterministic for debugging + mask[rnd_idx] = 1 + + num_g_real_masked = 0 + for g_idx in range(num_g_real): + if mask[g_idx] == 1: + masked_raw_cmc[num_g_real_masked] = raw_cmc[g_idx] + num_g_real_masked += 1 + + masked_cmc = np.zeros(num_g, dtype=np.float32) + function_cumsum(masked_raw_cmc, masked_cmc, num_g_real_masked) + for g_idx in range(num_g_real_masked): + if masked_cmc[g_idx] > 1: + masked_cmc[g_idx] = 1 + + for rank_idx in range(max_rank): + cmc[rank_idx] += masked_cmc[rank_idx] / num_repeats + + for rank_idx in range(max_rank): + all_cmc[q_idx, rank_idx] = cmc[rank_idx] + # compute average precision + # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision + function_cumsum(raw_cmc, tmp_cmc, num_g_real) + num_rel = 0 + tmp_cmc_sum = 0 + for g_idx in range(num_g_real): + tmp_cmc_sum += (tmp_cmc[g_idx] / (g_idx + 1.)) * raw_cmc[g_idx] + num_rel += raw_cmc[g_idx] + all_AP[q_idx] = tmp_cmc_sum / num_rel + num_valid_q += 1. + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + # compute averaged cmc + cdef float[:] avg_cmc = np.zeros(max_rank, dtype=np.float32) + for rank_idx in range(max_rank): + for q_idx in range(num_q): + avg_cmc[rank_idx] += all_cmc[q_idx, rank_idx] + avg_cmc[rank_idx] /= num_valid_q + + cdef float mAP = 0 + for q_idx in range(num_q): + mAP += all_AP[q_idx] + mAP /= num_valid_q + + return np.asarray(avg_cmc).astype(np.float32), mAP + + +cpdef eval_market1501_cy(float[:,:] distmat, int64_t[:] q_pids, int64_t[:]g_pids, + int64_t[:]q_camids, int64_t[:]g_camids, int64_t max_rank): + + cdef int64_t num_q = distmat.shape[0] + cdef int64_t num_g = distmat.shape[1] + + if num_g < max_rank: + max_rank = num_g + print('Note: number of gallery samples is quite small, got {}'.format(num_g)) + + cdef: + int64_t[:,:] indices = np.argsort(distmat, axis=1) + int64_t[:,:] matches = (np.asarray(g_pids)[np.asarray(indices)] == np.asarray(q_pids)[:, np.newaxis]).astype(np.int64) + + float[:,:] all_cmc = np.zeros((num_q, max_rank), dtype=np.float32) + float[:] all_AP = np.zeros(num_q, dtype=np.float32) + float num_valid_q = 0. # number of valid query + + int64_t q_idx, q_pid, q_camid, g_idx + int64_t[:] order = np.zeros(num_g, dtype=np.int64) + int64_t keep + + float[:] raw_cmc = np.zeros(num_g, dtype=np.float32) # binary vector, positions with value 1 are correct matches + float[:] cmc = np.zeros(num_g, dtype=np.float32) + int64_t num_g_real, rank_idx + uint64_t meet_condition + + float num_rel + float[:] tmp_cmc = np.zeros(num_g, dtype=np.float32) + float tmp_cmc_sum + + for q_idx in range(num_q): + # get query pid and camid + q_pid = q_pids[q_idx] + q_camid = q_camids[q_idx] + + # remove gallery samples that have the same pid and camid with query + for g_idx in range(num_g): + order[g_idx] = indices[q_idx, g_idx] + num_g_real = 0 + meet_condition = 0 + + for g_idx in range(num_g): + if (g_pids[order[g_idx]] != q_pid) or (g_camids[order[g_idx]] != q_camid): + raw_cmc[num_g_real] = matches[q_idx][g_idx] + num_g_real += 1 + if matches[q_idx][g_idx] > 1e-31: + meet_condition = 1 + + if not meet_condition: + # this condition is true when query identity does not appear in gallery + continue + + # compute cmc + function_cumsum(raw_cmc, cmc, num_g_real) + for g_idx in range(num_g_real): + if cmc[g_idx] > 1: + cmc[g_idx] = 1 + + for rank_idx in range(max_rank): + all_cmc[q_idx, rank_idx] = cmc[rank_idx] + num_valid_q += 1. + + # compute average precision + # reference: https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision + function_cumsum(raw_cmc, tmp_cmc, num_g_real) + num_rel = 0 + tmp_cmc_sum = 0 + for g_idx in range(num_g_real): + tmp_cmc_sum += (tmp_cmc[g_idx] / (g_idx + 1.)) * raw_cmc[g_idx] + num_rel += raw_cmc[g_idx] + all_AP[q_idx] = tmp_cmc_sum / num_rel + + assert num_valid_q > 0, 'Error: all query identities do not appear in gallery' + + # compute averaged cmc + cdef float[:] avg_cmc = np.zeros(max_rank, dtype=np.float32) + for rank_idx in range(max_rank): + for q_idx in range(num_q): + avg_cmc[rank_idx] += all_cmc[q_idx, rank_idx] + avg_cmc[rank_idx] /= num_valid_q + + cdef float mAP = 0 + for q_idx in range(num_q): + mAP += all_AP[q_idx] + mAP /= num_valid_q + + return np.asarray(avg_cmc).astype(np.float32), mAP + + +# Compute the cumulative sum +cdef void function_cumsum(cython.numeric[:] src, cython.numeric[:] dst, int64_t n): + cdef int64_t i + dst[0] = src[0] + for i in range(1, n): + dst[i] = src[i] + dst[i - 1] \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py new file mode 100644 index 0000000000..2c2d996fcc --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/setup.py @@ -0,0 +1,73 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +import numpy as np +from distutils.core import setup +from distutils.extension import Extension +from Cython.Build import cythonize + + +def numpy_include(): + try: + numpy_include = np.get_include() + except AttributeError: + numpy_include = np.get_numpy_include() + return numpy_include + + +ext_modules = [ + Extension( + 'rank_cy', + ['rank_cy.pyx'], + include_dirs=[numpy_include()], + ) +] + +setup( + name='Cython-based reid evaluation code', + ext_modules=cythonize(ext_modules) +) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py new file mode 100644 index 0000000000..67e5f61114 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/metrics/rank_cylib/test_cython.py @@ -0,0 +1,130 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function +import sys +import numpy as np +import timeit +import os.path as osp + +from torchreid import metrics + +sys.path.insert(0, osp.dirname(osp.abspath(__file__)) + '/../../..') +""" +Test the speed of cython-based evaluation code. The speed improvements +can be much bigger when using the real reid data, which contains a larger +amount of query and gallery images. + +Note: you might encounter the following error: + 'AssertionError: Error: all query identities do not appear in gallery'. +This is normal because the inputs are random numbers. Just try again. +""" + +print('*** Compare running time ***') + +setup = ''' +import sys +import os.path as osp +import numpy as np +sys.path.insert(0, osp.dirname(osp.abspath(__file__)) + '/../../..') +from torchreid import metrics +num_q = 30 +num_g = 300 +max_rank = 5 +distmat = np.random.rand(num_q, num_g) * 20 +q_pids = np.random.randint(0, num_q, size=num_q) +g_pids = np.random.randint(0, num_g, size=num_g) +q_camids = np.random.randint(0, 5, size=num_q) +g_camids = np.random.randint(0, 5, size=num_g) +''' + +print('=> Using market1501\'s metric') +pytime = timeit.timeit( + 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=False)', + setup=setup, + number=20 +) +cytime = timeit.timeit( + 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=True)', + setup=setup, + number=20 +) +print('Python time: {} s'.format(pytime)) +print('Cython time: {} s'.format(cytime)) +print('Cython is {} times faster than python\n'.format(pytime / cytime)) + +print('=> Using cuhk03\'s metric') +pytime = timeit.timeit( + 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=True, use_cython=False)', + setup=setup, + number=20 +) +cytime = timeit.timeit( + 'metrics.evaluate_rank(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_metric_cuhk03=True, use_cython=True)', + setup=setup, + number=20 +) +print('Python time: {} s'.format(pytime)) +print('Cython time: {} s'.format(cytime)) +print('Cython is {} times faster than python\n'.format(pytime / cytime)) +""" +print("=> Check precision") + +num_q = 30 +num_g = 300 +max_rank = 5 +distmat = np.random.rand(num_q, num_g) * 20 +q_pids = np.random.randint(0, num_q, size=num_q) +g_pids = np.random.randint(0, num_g, size=num_g) +q_camids = np.random.randint(0, 5, size=num_q) +g_camids = np.random.randint(0, 5, size=num_g) + +cmc, mAP = evaluate(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=False) +print("Python:\nmAP = {} \ncmc = {}\n".format(mAP, cmc)) +cmc, mAP = evaluate(distmat, q_pids, g_pids, q_camids, g_camids, max_rank, use_cython=True) +print("Cython:\nmAP = {} \ncmc = {}\n".format(mAP, cmc)) +""" diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py new file mode 100644 index 0000000000..7d0a09d988 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/__init__.py @@ -0,0 +1,166 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import +import torch + +from .pcb import * +from .mlfn import * +from .hacnn import * +from .osnet import * +from .senet import * +from .mudeep import * +from .nasnet import * +from .resnet import * +from .densenet import * +from .xception import * +from .osnet_ain import * +from .resnetmid import * +from .shufflenet import * +from .squeezenet import * +from .inceptionv4 import * +from .mobilenetv2 import * +from .resnet_ibn_a import * +from .resnet_ibn_b import * +from .shufflenetv2 import * +from .inceptionresnetv2 import * + +__model_factory = { + # image classification models + 'resnet18': resnet18, + 'resnet34': resnet34, + 'resnet50': resnet50, + 'resnet101': resnet101, + 'resnet152': resnet152, + 'resnext50_32x4d': resnext50_32x4d, + 'resnext101_32x8d': resnext101_32x8d, + 'resnet50_fc512': resnet50_fc512, + 'se_resnet50': se_resnet50, + 'se_resnet50_fc512': se_resnet50_fc512, + 'se_resnet101': se_resnet101, + 'se_resnext50_32x4d': se_resnext50_32x4d, + 'se_resnext101_32x4d': se_resnext101_32x4d, + 'densenet121': densenet121, + 'densenet169': densenet169, + 'densenet201': densenet201, + 'densenet161': densenet161, + 'densenet121_fc512': densenet121_fc512, + 'inceptionresnetv2': inceptionresnetv2, + 'inceptionv4': inceptionv4, + 'xception': xception, + 'resnet50_ibn_a': resnet50_ibn_a, + 'resnet50_ibn_b': resnet50_ibn_b, + # lightweight models + 'nasnsetmobile': nasnetamobile, + 'mobilenetv2_x1_0': mobilenetv2_x1_0, + 'mobilenetv2_x1_4': mobilenetv2_x1_4, + 'shufflenet': shufflenet, + 'squeezenet1_0': squeezenet1_0, + 'squeezenet1_0_fc512': squeezenet1_0_fc512, + 'squeezenet1_1': squeezenet1_1, + 'shufflenet_v2_x0_5': shufflenet_v2_x0_5, + 'shufflenet_v2_x1_0': shufflenet_v2_x1_0, + 'shufflenet_v2_x1_5': shufflenet_v2_x1_5, + 'shufflenet_v2_x2_0': shufflenet_v2_x2_0, + # reid-specific models + 'mudeep': MuDeep, + 'resnet50mid': resnet50mid, + 'hacnn': HACNN, + 'pcb_p6': pcb_p6, + 'pcb_p4': pcb_p4, + 'mlfn': mlfn, + 'osnet_x1_0': osnet_x1_0, + 'osnet_x0_75': osnet_x0_75, + 'osnet_x0_5': osnet_x0_5, + 'osnet_x0_25': osnet_x0_25, + 'osnet_ibn_x1_0': osnet_ibn_x1_0, + 'osnet_ain_x1_0': osnet_ain_x1_0 +} + + +def show_avai_models(): + """Displays available models. + + Examples:: + >>> from torchreid import models + >>> models.show_avai_models() + """ + print(list(__model_factory.keys())) + + +def build_model( + name, num_classes, loss='softmax', pretrained=True, use_gpu=True +): + """A function wrapper for building a model. + + Args: + name (str): model name. + num_classes (int): number of training identities. + loss (str, optional): loss function to optimize the model. Currently + supports "softmax" and "triplet". Default is "softmax". + pretrained (bool, optional): whether to load ImageNet-pretrained weights. + Default is True. + use_gpu (bool, optional): whether to use gpu. Default is True. + + Returns: + nn.Module + + Examples:: + >>> from torchreid import models + >>> model = models.build_model('resnet50', 751, loss='softmax') + """ + avai_models = list(__model_factory.keys()) + if name not in avai_models: + raise KeyError( + 'Unknown model: {}. Must be one of {}'.format(name, avai_models) + ) + return __model_factory[name]( + num_classes=num_classes, + loss=loss, + pretrained=pretrained, + use_gpu=use_gpu + ) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py new file mode 100644 index 0000000000..204aea72a1 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/densenet.py @@ -0,0 +1,425 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code source: https://github.com/pytorch/vision +""" +from __future__ import division, absolute_import +import re +from collections import OrderedDict +import torch +import torch.nn as nn +from torch.nn import functional as F +from torch.utils import model_zoo + +__all__ = [ + 'densenet121', 'densenet169', 'densenet201', 'densenet161', + 'densenet121_fc512' +] + +model_urls = { + 'densenet121': + 'https://download.pytorch.org/models/densenet121-a639ec97.pth', + 'densenet169': + 'https://download.pytorch.org/models/densenet169-b2777c0a.pth', + 'densenet201': + 'https://download.pytorch.org/models/densenet201-c1103571.pth', + 'densenet161': + 'https://download.pytorch.org/models/densenet161-8d451a50.pth', +} + + +class _DenseLayer(nn.Sequential): + + def __init__(self, num_input_features, growth_rate, bn_size, drop_rate): + super(_DenseLayer, self).__init__() + self.add_module('norm1', nn.BatchNorm2d(num_input_features)), + self.add_module('relu1', nn.ReLU(inplace=True)), + self.add_module( + 'conv1', + nn.Conv2d( + num_input_features, + bn_size * growth_rate, + kernel_size=1, + stride=1, + bias=False + ) + ), + self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate)), + self.add_module('relu2', nn.ReLU(inplace=True)), + self.add_module( + 'conv2', + nn.Conv2d( + bn_size * growth_rate, + growth_rate, + kernel_size=3, + stride=1, + padding=1, + bias=False + ) + ), + self.drop_rate = drop_rate + + def forward(self, x): + new_features = super(_DenseLayer, self).forward(x) + if self.drop_rate > 0: + new_features = F.dropout( + new_features, p=self.drop_rate, training=self.training + ) + return torch.cat([x, new_features], 1) + + +class _DenseBlock(nn.Sequential): + + def __init__( + self, num_layers, num_input_features, bn_size, growth_rate, drop_rate + ): + super(_DenseBlock, self).__init__() + for i in range(num_layers): + layer = _DenseLayer( + num_input_features + i*growth_rate, growth_rate, bn_size, + drop_rate + ) + self.add_module('denselayer%d' % (i+1), layer) + + +class _Transition(nn.Sequential): + + def __init__(self, num_input_features, num_output_features): + super(_Transition, self).__init__() + self.add_module('norm', nn.BatchNorm2d(num_input_features)) + self.add_module('relu', nn.ReLU(inplace=True)) + self.add_module( + 'conv', + nn.Conv2d( + num_input_features, + num_output_features, + kernel_size=1, + stride=1, + bias=False + ) + ) + self.add_module('pool', nn.AvgPool2d(kernel_size=2, stride=2)) + + +class DenseNet(nn.Module): + """Densely connected network. + + Reference: + Huang et al. Densely Connected Convolutional Networks. CVPR 2017. + + Public keys: + - ``densenet121``: DenseNet121. + - ``densenet169``: DenseNet169. + - ``densenet201``: DenseNet201. + - ``densenet161``: DenseNet161. + - ``densenet121_fc512``: DenseNet121 + FC. + """ + + def __init__( + self, + num_classes, + loss, + growth_rate=32, + block_config=(6, 12, 24, 16), + num_init_features=64, + bn_size=4, + drop_rate=0, + fc_dims=None, + dropout_p=None, + **kwargs + ): + + super(DenseNet, self).__init__() + self.loss = loss + + # First convolution + self.features = nn.Sequential( + OrderedDict( + [ + ( + 'conv0', + nn.Conv2d( + 3, + num_init_features, + kernel_size=7, + stride=2, + padding=3, + bias=False + ) + ), + ('norm0', nn.BatchNorm2d(num_init_features)), + ('relu0', nn.ReLU(inplace=True)), + ( + 'pool0', + nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + ), + ] + ) + ) + + # Each denseblock + num_features = num_init_features + for i, num_layers in enumerate(block_config): + block = _DenseBlock( + num_layers=num_layers, + num_input_features=num_features, + bn_size=bn_size, + growth_rate=growth_rate, + drop_rate=drop_rate + ) + self.features.add_module('denseblock%d' % (i+1), block) + num_features = num_features + num_layers*growth_rate + if i != len(block_config) - 1: + trans = _Transition( + num_input_features=num_features, + num_output_features=num_features // 2 + ) + self.features.add_module('transition%d' % (i+1), trans) + num_features = num_features // 2 + + # Final batch norm + self.features.add_module('norm5', nn.BatchNorm2d(num_features)) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.feature_dim = num_features + self.fc = self._construct_fc_layer(fc_dims, num_features, dropout_p) + + # Linear layer + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer. + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def forward(self, x): + f = self.features(x) + f = F.relu(f, inplace=True) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + + # '.'s are no longer allowed in module names, but pervious _DenseLayer + # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. + # They are also in the checkpoints in model_urls. This pattern is used + # to find such keys. + pattern = re.compile( + r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$' + ) + for key in list(pretrain_dict.keys()): + res = pattern.match(key) + if res: + new_key = res.group(1) + res.group(2) + pretrain_dict[new_key] = pretrain_dict[key] + del pretrain_dict[key] + + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +""" +Dense network configurations: +-- +densenet121: num_init_features=64, growth_rate=32, block_config=(6, 12, 24, 16) +densenet169: num_init_features=64, growth_rate=32, block_config=(6, 12, 32, 32) +densenet201: num_init_features=64, growth_rate=32, block_config=(6, 12, 48, 32) +densenet161: num_init_features=96, growth_rate=48, block_config=(6, 12, 36, 24) +""" + + +def densenet121(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 24, 16), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet121']) + return model + + +def densenet169(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 32, 32), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet169']) + return model + + +def densenet201(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 48, 32), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet201']) + return model + + +def densenet161(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=96, + growth_rate=48, + block_config=(6, 12, 36, 24), + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet161']) + return model + + +def densenet121_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): + model = DenseNet( + num_classes=num_classes, + loss=loss, + num_init_features=64, + growth_rate=32, + block_config=(6, 12, 24, 16), + fc_dims=[512], + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['densenet121']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py new file mode 100644 index 0000000000..27dae2fa28 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/hacnn.py @@ -0,0 +1,461 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +from torch import nn +from torch.nn import functional as F + +__all__ = ['HACNN'] + + +class ConvBlock(nn.Module): + """Basic convolutional block. + + convolution + batch normalization + relu. + + Args: + in_c (int): number of input channels. + out_c (int): number of output channels. + k (int or tuple): kernel size. + s (int or tuple): stride. + p (int or tuple): padding. + """ + + def __init__(self, in_c, out_c, k, s=1, p=0): + super(ConvBlock, self).__init__() + self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) + self.bn = nn.BatchNorm2d(out_c) + + def forward(self, x): + return F.relu(self.bn(self.conv(x))) + + +class InceptionA(nn.Module): + + def __init__(self, in_channels, out_channels): + super(InceptionA, self).__init__() + mid_channels = out_channels // 4 + + self.stream1 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream2 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream3 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ) + self.stream4 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1), + ConvBlock(in_channels, mid_channels, 1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + s4 = self.stream4(x) + y = torch.cat([s1, s2, s3, s4], dim=1) + return y + + +class InceptionB(nn.Module): + + def __init__(self, in_channels, out_channels): + super(InceptionB, self).__init__() + mid_channels = out_channels // 4 + + self.stream1 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), + ) + self.stream2 = nn.Sequential( + ConvBlock(in_channels, mid_channels, 1), + ConvBlock(mid_channels, mid_channels, 3, p=1), + ConvBlock(mid_channels, mid_channels, 3, s=2, p=1), + ) + self.stream3 = nn.Sequential( + nn.MaxPool2d(3, stride=2, padding=1), + ConvBlock(in_channels, mid_channels * 2, 1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + y = torch.cat([s1, s2, s3], dim=1) + return y + + +class SpatialAttn(nn.Module): + """Spatial Attention (Sec. 3.1.I.1)""" + + def __init__(self): + super(SpatialAttn, self).__init__() + self.conv1 = ConvBlock(1, 1, 3, s=2, p=1) + self.conv2 = ConvBlock(1, 1, 1) + + def forward(self, x): + # global cross-channel averaging + x = x.mean(1, keepdim=True) + # 3-by-3 conv + x = self.conv1(x) + # bilinear resizing + x = F.upsample( + x, (x.size(2) * 2, x.size(3) * 2), + mode='bilinear', + align_corners=True + ) + # scaling conv + x = self.conv2(x) + return x + + +class ChannelAttn(nn.Module): + """Channel Attention (Sec. 3.1.I.2)""" + + def __init__(self, in_channels, reduction_rate=16): + super(ChannelAttn, self).__init__() + assert in_channels % reduction_rate == 0 + self.conv1 = ConvBlock(in_channels, in_channels // reduction_rate, 1) + self.conv2 = ConvBlock(in_channels // reduction_rate, in_channels, 1) + + def forward(self, x): + # squeeze operation (global average pooling) + x = F.avg_pool2d(x, x.size()[2:]) + # excitation operation (2 conv layers) + x = self.conv1(x) + x = self.conv2(x) + return x + + +class SoftAttn(nn.Module): + """Soft Attention (Sec. 3.1.I) + + Aim: Spatial Attention + Channel Attention + + Output: attention maps with shape identical to input. + """ + + def __init__(self, in_channels): + super(SoftAttn, self).__init__() + self.spatial_attn = SpatialAttn() + self.channel_attn = ChannelAttn(in_channels) + self.conv = ConvBlock(in_channels, in_channels, 1) + + def forward(self, x): + y_spatial = self.spatial_attn(x) + y_channel = self.channel_attn(x) + y = y_spatial * y_channel + y = torch.sigmoid(self.conv(y)) + return y + + +class HardAttn(nn.Module): + """Hard Attention (Sec. 3.1.II)""" + + def __init__(self, in_channels): + super(HardAttn, self).__init__() + self.fc = nn.Linear(in_channels, 4 * 2) + self.init_params() + + def init_params(self): + self.fc.weight.data.zero_() + self.fc.bias.data.copy_( + torch.tensor( + [0, -0.75, 0, -0.25, 0, 0.25, 0, 0.75], dtype=torch.float + ) + ) + + def forward(self, x): + # squeeze operation (global average pooling) + x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), x.size(1)) + # predict transformation parameters + theta = torch.tanh(self.fc(x)) + theta = theta.view(-1, 4, 2) + return theta + + +class HarmAttn(nn.Module): + """Harmonious Attention (Sec. 3.1)""" + + def __init__(self, in_channels): + super(HarmAttn, self).__init__() + self.soft_attn = SoftAttn(in_channels) + self.hard_attn = HardAttn(in_channels) + + def forward(self, x): + y_soft_attn = self.soft_attn(x) + theta = self.hard_attn(x) + return y_soft_attn, theta + + +class HACNN(nn.Module): + """Harmonious Attention Convolutional Neural Network. + + Reference: + Li et al. Harmonious Attention Network for Person Re-identification. CVPR 2018. + + Public keys: + - ``hacnn``: HACNN. + """ + + # Args: + # num_classes (int): number of classes to predict + # nchannels (list): number of channels AFTER concatenation + # feat_dim (int): feature dimension for a single stream + # learn_region (bool): whether to learn region features (i.e. local branch) + + def __init__( + self, + num_classes, + loss='softmax', + nchannels=[128, 256, 384], + feat_dim=512, + learn_region=True, + use_gpu=True, + **kwargs + ): + super(HACNN, self).__init__() + self.loss = loss + self.learn_region = learn_region + self.use_gpu = use_gpu + + self.conv = ConvBlock(3, 32, 3, s=2, p=1) + + # Construct Inception + HarmAttn blocks + # ============== Block 1 ============== + self.inception1 = nn.Sequential( + InceptionA(32, nchannels[0]), + InceptionB(nchannels[0], nchannels[0]), + ) + self.ha1 = HarmAttn(nchannels[0]) + + # ============== Block 2 ============== + self.inception2 = nn.Sequential( + InceptionA(nchannels[0], nchannels[1]), + InceptionB(nchannels[1], nchannels[1]), + ) + self.ha2 = HarmAttn(nchannels[1]) + + # ============== Block 3 ============== + self.inception3 = nn.Sequential( + InceptionA(nchannels[1], nchannels[2]), + InceptionB(nchannels[2], nchannels[2]), + ) + self.ha3 = HarmAttn(nchannels[2]) + + self.fc_global = nn.Sequential( + nn.Linear(nchannels[2], feat_dim), + nn.BatchNorm1d(feat_dim), + nn.ReLU(), + ) + self.classifier_global = nn.Linear(feat_dim, num_classes) + + if self.learn_region: + self.init_scale_factors() + self.local_conv1 = InceptionB(32, nchannels[0]) + self.local_conv2 = InceptionB(nchannels[0], nchannels[1]) + self.local_conv3 = InceptionB(nchannels[1], nchannels[2]) + self.fc_local = nn.Sequential( + nn.Linear(nchannels[2] * 4, feat_dim), + nn.BatchNorm1d(feat_dim), + nn.ReLU(), + ) + self.classifier_local = nn.Linear(feat_dim, num_classes) + self.feat_dim = feat_dim * 2 + else: + self.feat_dim = feat_dim + + def init_scale_factors(self): + # initialize scale factors (s_w, s_h) for four regions + self.scale_factors = [] + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + self.scale_factors.append( + torch.tensor([[1, 0], [0, 0.25]], dtype=torch.float) + ) + + def stn(self, x, theta): + """Performs spatial transform + + x: (batch, channel, height, width) + theta: (batch, 2, 3) + """ + grid = F.affine_grid(theta, x.size()) + x = F.grid_sample(x, grid) + return x + + def transform_theta(self, theta_i, region_idx): + """Transforms theta to include (s_w, s_h), resulting in (batch, 2, 3)""" + scale_factors = self.scale_factors[region_idx] + theta = torch.zeros(theta_i.size(0), 2, 3) + theta[:, :, :2] = scale_factors + theta[:, :, -1] = theta_i + if self.use_gpu: + theta = theta.cuda() + return theta + + def forward(self, x): + assert x.size(2) == 160 and x.size(3) == 64, \ + 'Input size does not match, expected (160, 64) but got ({}, {})'.format(x.size(2), x.size(3)) + x = self.conv(x) + + # ============== Block 1 ============== + # global branch + x1 = self.inception1(x) + x1_attn, x1_theta = self.ha1(x1) + x1_out = x1 * x1_attn + # local branch + if self.learn_region: + x1_local_list = [] + for region_idx in range(4): + x1_theta_i = x1_theta[:, region_idx, :] + x1_theta_i = self.transform_theta(x1_theta_i, region_idx) + x1_trans_i = self.stn(x, x1_theta_i) + x1_trans_i = F.upsample( + x1_trans_i, (24, 28), mode='bilinear', align_corners=True + ) + x1_local_i = self.local_conv1(x1_trans_i) + x1_local_list.append(x1_local_i) + + # ============== Block 2 ============== + # Block 2 + # global branch + x2 = self.inception2(x1_out) + x2_attn, x2_theta = self.ha2(x2) + x2_out = x2 * x2_attn + # local branch + if self.learn_region: + x2_local_list = [] + for region_idx in range(4): + x2_theta_i = x2_theta[:, region_idx, :] + x2_theta_i = self.transform_theta(x2_theta_i, region_idx) + x2_trans_i = self.stn(x1_out, x2_theta_i) + x2_trans_i = F.upsample( + x2_trans_i, (12, 14), mode='bilinear', align_corners=True + ) + x2_local_i = x2_trans_i + x1_local_list[region_idx] + x2_local_i = self.local_conv2(x2_local_i) + x2_local_list.append(x2_local_i) + + # ============== Block 3 ============== + # Block 3 + # global branch + x3 = self.inception3(x2_out) + x3_attn, x3_theta = self.ha3(x3) + x3_out = x3 * x3_attn + # local branch + if self.learn_region: + x3_local_list = [] + for region_idx in range(4): + x3_theta_i = x3_theta[:, region_idx, :] + x3_theta_i = self.transform_theta(x3_theta_i, region_idx) + x3_trans_i = self.stn(x2_out, x3_theta_i) + x3_trans_i = F.upsample( + x3_trans_i, (6, 7), mode='bilinear', align_corners=True + ) + x3_local_i = x3_trans_i + x2_local_list[region_idx] + x3_local_i = self.local_conv3(x3_local_i) + x3_local_list.append(x3_local_i) + + # ============== Feature generation ============== + # global branch + x_global = F.avg_pool2d(x3_out, + x3_out.size()[2:] + ).view(x3_out.size(0), x3_out.size(1)) + x_global = self.fc_global(x_global) + # local branch + if self.learn_region: + x_local_list = [] + for region_idx in range(4): + x_local_i = x3_local_list[region_idx] + x_local_i = F.avg_pool2d(x_local_i, + x_local_i.size()[2:] + ).view(x_local_i.size(0), -1) + x_local_list.append(x_local_i) + x_local = torch.cat(x_local_list, 1) + x_local = self.fc_local(x_local) + + if not self.training: + # l2 normalization before concatenation + if self.learn_region: + x_global = x_global / x_global.norm(p=2, dim=1, keepdim=True) + x_local = x_local / x_local.norm(p=2, dim=1, keepdim=True) + return torch.cat([x_global, x_local], 1) + else: + return x_global + + prelogits_global = self.classifier_global(x_global) + if self.learn_region: + prelogits_local = self.classifier_local(x_local) + + if self.loss == 'softmax': + if self.learn_region: + return (prelogits_global, prelogits_local) + else: + return prelogits_global + + elif self.loss == 'triplet': + if self.learn_region: + return (prelogits_global, prelogits_local), (x_global, x_local) + else: + return prelogits_global, x_global + + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py new file mode 100644 index 0000000000..f9d62718b3 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionresnetv2.py @@ -0,0 +1,406 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['inceptionresnetv2'] + +pretrained_settings = { + 'inceptionresnetv2': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000 + }, + 'imagenet+background': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionresnetv2-520b38e4.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1001 + } + } +} + + +class BasicConv2d(nn.Module): + + def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): + super(BasicConv2d, self).__init__() + self.conv = nn.Conv2d( + in_planes, + out_planes, + kernel_size=kernel_size, + stride=stride, + padding=padding, + bias=False + ) # verify bias false + self.bn = nn.BatchNorm2d( + out_planes, + eps=0.001, # value found in tensorflow + momentum=0.1, # default pytorch value + affine=True + ) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Mixed_5b(nn.Module): + + def __init__(self): + super(Mixed_5b, self).__init__() + + self.branch0 = BasicConv2d(192, 96, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(192, 48, kernel_size=1, stride=1), + BasicConv2d(48, 64, kernel_size=5, stride=1, padding=2) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(192, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), + BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(192, 64, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Block35(nn.Module): + + def __init__(self, scale=1.0): + super(Block35, self).__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(320, 32, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(320, 32, kernel_size=1, stride=1), + BasicConv2d(32, 32, kernel_size=3, stride=1, padding=1) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(320, 32, kernel_size=1, stride=1), + BasicConv2d(32, 48, kernel_size=3, stride=1, padding=1), + BasicConv2d(48, 64, kernel_size=3, stride=1, padding=1) + ) + + self.conv2d = nn.Conv2d(128, 320, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Mixed_6a(nn.Module): + + def __init__(self): + super(Mixed_6a, self).__init__() + + self.branch0 = BasicConv2d(320, 384, kernel_size=3, stride=2) + + self.branch1 = nn.Sequential( + BasicConv2d(320, 256, kernel_size=1, stride=1), + BasicConv2d(256, 256, kernel_size=3, stride=1, padding=1), + BasicConv2d(256, 384, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Block17(nn.Module): + + def __init__(self, scale=1.0): + super(Block17, self).__init__() + + self.scale = scale + + self.branch0 = BasicConv2d(1088, 192, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(1088, 128, kernel_size=1, stride=1), + BasicConv2d( + 128, 160, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 160, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) + ) + ) + + self.conv2d = nn.Conv2d(384, 1088, kernel_size=1, stride=1) + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + out = self.relu(out) + return out + + +class Mixed_7a(nn.Module): + + def __init__(self): + super(Mixed_7a, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 384, kernel_size=3, stride=2) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 288, kernel_size=3, stride=2) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(1088, 256, kernel_size=1, stride=1), + BasicConv2d(256, 288, kernel_size=3, stride=1, padding=1), + BasicConv2d(288, 320, kernel_size=3, stride=2) + ) + + self.branch3 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Block8(nn.Module): + + def __init__(self, scale=1.0, noReLU=False): + super(Block8, self).__init__() + + self.scale = scale + self.noReLU = noReLU + + self.branch0 = BasicConv2d(2080, 192, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(2080, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 224, kernel_size=(1, 3), stride=1, padding=(0, 1) + ), + BasicConv2d( + 224, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + ) + + self.conv2d = nn.Conv2d(448, 2080, kernel_size=1, stride=1) + if not self.noReLU: + self.relu = nn.ReLU(inplace=False) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + out = self.conv2d(out) + out = out * self.scale + x + if not self.noReLU: + out = self.relu(out) + return out + + +# ---------------- +# Model Definition +# ---------------- +class InceptionResNetV2(nn.Module): + """Inception-ResNet-V2. + + Reference: + Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual + Connections on Learning. AAAI 2017. + + Public keys: + - ``inceptionresnetv2``: Inception-ResNet-V2. + """ + + def __init__(self, num_classes, loss='softmax', **kwargs): + super(InceptionResNetV2, self).__init__() + self.loss = loss + + # Modules + self.conv2d_1a = BasicConv2d(3, 32, kernel_size=3, stride=2) + self.conv2d_2a = BasicConv2d(32, 32, kernel_size=3, stride=1) + self.conv2d_2b = BasicConv2d( + 32, 64, kernel_size=3, stride=1, padding=1 + ) + self.maxpool_3a = nn.MaxPool2d(3, stride=2) + self.conv2d_3b = BasicConv2d(64, 80, kernel_size=1, stride=1) + self.conv2d_4a = BasicConv2d(80, 192, kernel_size=3, stride=1) + self.maxpool_5a = nn.MaxPool2d(3, stride=2) + self.mixed_5b = Mixed_5b() + self.repeat = nn.Sequential( + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17), Block35(scale=0.17), Block35(scale=0.17), + Block35(scale=0.17) + ) + self.mixed_6a = Mixed_6a() + self.repeat_1 = nn.Sequential( + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10), Block17(scale=0.10), + Block17(scale=0.10), Block17(scale=0.10) + ) + self.mixed_7a = Mixed_7a() + self.repeat_2 = nn.Sequential( + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20), + Block8(scale=0.20), Block8(scale=0.20), Block8(scale=0.20) + ) + + self.block8 = Block8(noReLU=True) + self.conv2d_7b = BasicConv2d(2080, 1536, kernel_size=1, stride=1) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.classifier = nn.Linear(1536, num_classes) + + def load_imagenet_weights(self): + settings = pretrained_settings['inceptionresnetv2']['imagenet'] + pretrain_dict = model_zoo.load_url(settings['url']) + model_dict = self.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + self.load_state_dict(model_dict) + + def featuremaps(self, x): + x = self.conv2d_1a(x) + x = self.conv2d_2a(x) + x = self.conv2d_2b(x) + x = self.maxpool_3a(x) + x = self.conv2d_3b(x) + x = self.conv2d_4a(x) + x = self.maxpool_5a(x) + x = self.mixed_5b(x) + x = self.repeat(x) + x = self.mixed_6a(x) + x = self.repeat_1(x) + x = self.mixed_7a(x) + x = self.repeat_2(x) + x = self.block8(x) + x = self.conv2d_7b(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def inceptionresnetv2(num_classes, loss='softmax', pretrained=True, **kwargs): + model = InceptionResNetV2(num_classes=num_classes, loss=loss, **kwargs) + if pretrained: + model.load_imagenet_weights() + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py new file mode 100644 index 0000000000..32a847a88f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/inceptionv4.py @@ -0,0 +1,428 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['inceptionv4'] +""" +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" + +pretrained_settings = { + 'inceptionv4': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000 + }, + 'imagenet+background': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/inceptionv4-8e4777a0.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1001 + } + } +} + + +class BasicConv2d(nn.Module): + + def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0): + super(BasicConv2d, self).__init__() + self.conv = nn.Conv2d( + in_planes, + out_planes, + kernel_size=kernel_size, + stride=stride, + padding=padding, + bias=False + ) # verify bias false + self.bn = nn.BatchNorm2d( + out_planes, + eps=0.001, # value found in tensorflow + momentum=0.1, # default pytorch value + affine=True + ) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Mixed_3a(nn.Module): + + def __init__(self): + super(Mixed_3a, self).__init__() + self.maxpool = nn.MaxPool2d(3, stride=2) + self.conv = BasicConv2d(64, 96, kernel_size=3, stride=2) + + def forward(self, x): + x0 = self.maxpool(x) + x1 = self.conv(x) + out = torch.cat((x0, x1), 1) + return out + + +class Mixed_4a(nn.Module): + + def __init__(self): + super(Mixed_4a, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(160, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(160, 64, kernel_size=1, stride=1), + BasicConv2d(64, 64, kernel_size=(1, 7), stride=1, padding=(0, 3)), + BasicConv2d(64, 64, kernel_size=(7, 1), stride=1, padding=(3, 0)), + BasicConv2d(64, 96, kernel_size=(3, 3), stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + out = torch.cat((x0, x1), 1) + return out + + +class Mixed_5a(nn.Module): + + def __init__(self): + super(Mixed_5a, self).__init__() + self.conv = BasicConv2d(192, 192, kernel_size=3, stride=2) + self.maxpool = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.conv(x) + x1 = self.maxpool(x) + out = torch.cat((x0, x1), 1) + return out + + +class Inception_A(nn.Module): + + def __init__(self): + super(Inception_A, self).__init__() + self.branch0 = BasicConv2d(384, 96, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(384, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(384, 64, kernel_size=1, stride=1), + BasicConv2d(64, 96, kernel_size=3, stride=1, padding=1), + BasicConv2d(96, 96, kernel_size=3, stride=1, padding=1) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(384, 96, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Reduction_A(nn.Module): + + def __init__(self): + super(Reduction_A, self).__init__() + self.branch0 = BasicConv2d(384, 384, kernel_size=3, stride=2) + + self.branch1 = nn.Sequential( + BasicConv2d(384, 192, kernel_size=1, stride=1), + BasicConv2d(192, 224, kernel_size=3, stride=1, padding=1), + BasicConv2d(224, 256, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Inception_B(nn.Module): + + def __init__(self): + super(Inception_B, self).__init__() + self.branch0 = BasicConv2d(1024, 384, kernel_size=1, stride=1) + + self.branch1 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 224, 256, kernel_size=(7, 1), stride=1, padding=(3, 0) + ) + ) + + self.branch2 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d( + 192, 192, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), + BasicConv2d( + 192, 224, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 224, 224, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), + BasicConv2d( + 224, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) + ) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(1024, 128, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + x3 = self.branch3(x) + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class Reduction_B(nn.Module): + + def __init__(self): + super(Reduction_B, self).__init__() + + self.branch0 = nn.Sequential( + BasicConv2d(1024, 192, kernel_size=1, stride=1), + BasicConv2d(192, 192, kernel_size=3, stride=2) + ) + + self.branch1 = nn.Sequential( + BasicConv2d(1024, 256, kernel_size=1, stride=1), + BasicConv2d( + 256, 256, kernel_size=(1, 7), stride=1, padding=(0, 3) + ), + BasicConv2d( + 256, 320, kernel_size=(7, 1), stride=1, padding=(3, 0) + ), BasicConv2d(320, 320, kernel_size=3, stride=2) + ) + + self.branch2 = nn.MaxPool2d(3, stride=2) + + def forward(self, x): + x0 = self.branch0(x) + x1 = self.branch1(x) + x2 = self.branch2(x) + out = torch.cat((x0, x1, x2), 1) + return out + + +class Inception_C(nn.Module): + + def __init__(self): + super(Inception_C, self).__init__() + + self.branch0 = BasicConv2d(1536, 256, kernel_size=1, stride=1) + + self.branch1_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) + self.branch1_1a = BasicConv2d( + 384, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch1_1b = BasicConv2d( + 384, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + + self.branch2_0 = BasicConv2d(1536, 384, kernel_size=1, stride=1) + self.branch2_1 = BasicConv2d( + 384, 448, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + self.branch2_2 = BasicConv2d( + 448, 512, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch2_3a = BasicConv2d( + 512, 256, kernel_size=(1, 3), stride=1, padding=(0, 1) + ) + self.branch2_3b = BasicConv2d( + 512, 256, kernel_size=(3, 1), stride=1, padding=(1, 0) + ) + + self.branch3 = nn.Sequential( + nn.AvgPool2d(3, stride=1, padding=1, count_include_pad=False), + BasicConv2d(1536, 256, kernel_size=1, stride=1) + ) + + def forward(self, x): + x0 = self.branch0(x) + + x1_0 = self.branch1_0(x) + x1_1a = self.branch1_1a(x1_0) + x1_1b = self.branch1_1b(x1_0) + x1 = torch.cat((x1_1a, x1_1b), 1) + + x2_0 = self.branch2_0(x) + x2_1 = self.branch2_1(x2_0) + x2_2 = self.branch2_2(x2_1) + x2_3a = self.branch2_3a(x2_2) + x2_3b = self.branch2_3b(x2_2) + x2 = torch.cat((x2_3a, x2_3b), 1) + + x3 = self.branch3(x) + + out = torch.cat((x0, x1, x2, x3), 1) + return out + + +class InceptionV4(nn.Module): + """Inception-v4. + + Reference: + Szegedy et al. Inception-v4, Inception-ResNet and the Impact of Residual + Connections on Learning. AAAI 2017. + + Public keys: + - ``inceptionv4``: InceptionV4. + """ + + def __init__(self, num_classes, loss, **kwargs): + super(InceptionV4, self).__init__() + self.loss = loss + + self.features = nn.Sequential( + BasicConv2d(3, 32, kernel_size=3, stride=2), + BasicConv2d(32, 32, kernel_size=3, stride=1), + BasicConv2d(32, 64, kernel_size=3, stride=1, padding=1), + Mixed_3a(), + Mixed_4a(), + Mixed_5a(), + Inception_A(), + Inception_A(), + Inception_A(), + Inception_A(), + Reduction_A(), # Mixed_6a + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Inception_B(), + Reduction_B(), # Mixed_7a + Inception_C(), + Inception_C(), + Inception_C() + ) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.classifier = nn.Linear(1536, num_classes) + + def forward(self, x): + f = self.features(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def inceptionv4(num_classes, loss='softmax', pretrained=True, **kwargs): + model = InceptionV4(num_classes, loss, **kwargs) + if pretrained: + model_url = pretrained_settings['inceptionv4']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py new file mode 100644 index 0000000000..0e538241f7 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mlfn.py @@ -0,0 +1,316 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['mlfn'] + +model_urls = { + # training epoch = 5, top1 = 51.6 + 'imagenet': + 'https://mega.nz/#!YHxAhaxC!yu9E6zWl0x5zscSouTdbZu8gdFFytDdl-RAdD2DEfpk', +} + + +class MLFNBlock(nn.Module): + + def __init__( + self, in_channels, out_channels, stride, fsm_channels, groups=32 + ): + super(MLFNBlock, self).__init__() + self.groups = groups + mid_channels = out_channels // 2 + + # Factor Modules + self.fm_conv1 = nn.Conv2d(in_channels, mid_channels, 1, bias=False) + self.fm_bn1 = nn.BatchNorm2d(mid_channels) + self.fm_conv2 = nn.Conv2d( + mid_channels, + mid_channels, + 3, + stride=stride, + padding=1, + bias=False, + groups=self.groups + ) + self.fm_bn2 = nn.BatchNorm2d(mid_channels) + self.fm_conv3 = nn.Conv2d(mid_channels, out_channels, 1, bias=False) + self.fm_bn3 = nn.BatchNorm2d(out_channels) + + # Factor Selection Module + self.fsm = nn.Sequential( + nn.AdaptiveAvgPool2d(1), + nn.Conv2d(in_channels, fsm_channels[0], 1), + nn.BatchNorm2d(fsm_channels[0]), + nn.ReLU(inplace=True), + nn.Conv2d(fsm_channels[0], fsm_channels[1], 1), + nn.BatchNorm2d(fsm_channels[1]), + nn.ReLU(inplace=True), + nn.Conv2d(fsm_channels[1], self.groups, 1), + nn.BatchNorm2d(self.groups), + nn.Sigmoid(), + ) + + self.downsample = None + if in_channels != out_channels or stride > 1: + self.downsample = nn.Sequential( + nn.Conv2d( + in_channels, out_channels, 1, stride=stride, bias=False + ), + nn.BatchNorm2d(out_channels), + ) + + def forward(self, x): + residual = x + s = self.fsm(x) + + # reduce dimension + x = self.fm_conv1(x) + x = self.fm_bn1(x) + x = F.relu(x, inplace=True) + + # group convolution + x = self.fm_conv2(x) + x = self.fm_bn2(x) + x = F.relu(x, inplace=True) + + # factor selection + b, c = x.size(0), x.size(1) + n = c // self.groups + ss = s.repeat(1, n, 1, 1) # from (b, g, 1, 1) to (b, g*n=c, 1, 1) + ss = ss.view(b, n, self.groups, 1, 1) + ss = ss.permute(0, 2, 1, 3, 4).contiguous() + ss = ss.view(b, c, 1, 1) + x = ss * x + + # recover dimension + x = self.fm_conv3(x) + x = self.fm_bn3(x) + x = F.relu(x, inplace=True) + + if self.downsample is not None: + residual = self.downsample(residual) + + return F.relu(residual + x, inplace=True), s + + +class MLFN(nn.Module): + """Multi-Level Factorisation Net. + + Reference: + Chang et al. Multi-Level Factorisation Net for + Person Re-Identification. CVPR 2018. + + Public keys: + - ``mlfn``: MLFN (Multi-Level Factorisation Net). + """ + + def __init__( + self, + num_classes, + loss='softmax', + groups=32, + channels=[64, 256, 512, 1024, 2048], + embed_dim=1024, + **kwargs + ): + super(MLFN, self).__init__() + self.loss = loss + self.groups = groups + + # first convolutional layer + self.conv1 = nn.Conv2d(3, channels[0], 7, stride=2, padding=3) + self.bn1 = nn.BatchNorm2d(channels[0]) + self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) + + # main body + self.feature = nn.ModuleList( + [ + # layer 1-3 + MLFNBlock(channels[0], channels[1], 1, [128, 64], self.groups), + MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), + MLFNBlock(channels[1], channels[1], 1, [128, 64], self.groups), + # layer 4-7 + MLFNBlock( + channels[1], channels[2], 2, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + MLFNBlock( + channels[2], channels[2], 1, [256, 128], self.groups + ), + # layer 8-13 + MLFNBlock( + channels[2], channels[3], 2, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[3], channels[3], 1, [512, 128], self.groups + ), + # layer 14-16 + MLFNBlock( + channels[3], channels[4], 2, [512, 128], self.groups + ), + MLFNBlock( + channels[4], channels[4], 1, [512, 128], self.groups + ), + MLFNBlock( + channels[4], channels[4], 1, [512, 128], self.groups + ), + ] + ) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + + # projection functions + self.fc_x = nn.Sequential( + nn.Conv2d(channels[4], embed_dim, 1, bias=False), + nn.BatchNorm2d(embed_dim), + nn.ReLU(inplace=True), + ) + self.fc_s = nn.Sequential( + nn.Conv2d(self.groups * 16, embed_dim, 1, bias=False), + nn.BatchNorm2d(embed_dim), + nn.ReLU(inplace=True), + ) + + self.classifier = nn.Linear(embed_dim, num_classes) + + self.init_params() + + def init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def forward(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = F.relu(x, inplace=True) + x = self.maxpool(x) + + s_hat = [] + for block in self.feature: + x, s = block(x) + s_hat.append(s) + s_hat = torch.cat(s_hat, 1) + + x = self.global_avgpool(x) + x = self.fc_x(x) + s_hat = self.fc_s(s_hat) + + v = (x+s_hat) * 0.5 + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def mlfn(num_classes, loss='softmax', pretrained=True, **kwargs): + model = MLFN(num_classes, loss, **kwargs) + if pretrained: + # init_pretrained_weights(model, model_urls['imagenet']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['imagenet']) + ) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py new file mode 100644 index 0000000000..690dade1bb --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mobilenetv2.py @@ -0,0 +1,321 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['mobilenetv2_x1_0', 'mobilenetv2_x1_4'] + +model_urls = { + # 1.0: top-1 71.3 + 'mobilenetv2_x1_0': + 'https://mega.nz/#!NKp2wAIA!1NH1pbNzY_M2hVk_hdsxNM1NUOWvvGPHhaNr-fASF6c', + # 1.4: top-1 73.9 + 'mobilenetv2_x1_4': + 'https://mega.nz/#!RGhgEIwS!xN2s2ZdyqI6vQ3EwgmRXLEW3khr9tpXg96G9SUJugGk', +} + + +class ConvBlock(nn.Module): + """Basic convolutional block. + + convolution (bias discarded) + batch normalization + relu6. + + Args: + in_c (int): number of input channels. + out_c (int): number of output channels. + k (int or tuple): kernel size. + s (int or tuple): stride. + p (int or tuple): padding. + g (int): number of blocked connections from input channels + to output channels (default: 1). + """ + + def __init__(self, in_c, out_c, k, s=1, p=0, g=1): + super(ConvBlock, self).__init__() + self.conv = nn.Conv2d( + in_c, out_c, k, stride=s, padding=p, bias=False, groups=g + ) + self.bn = nn.BatchNorm2d(out_c) + + def forward(self, x): + return F.relu6(self.bn(self.conv(x))) + + +class Bottleneck(nn.Module): + + def __init__(self, in_channels, out_channels, expansion_factor, stride=1): + super(Bottleneck, self).__init__() + mid_channels = in_channels * expansion_factor + self.use_residual = stride == 1 and in_channels == out_channels + self.conv1 = ConvBlock(in_channels, mid_channels, 1) + self.dwconv2 = ConvBlock( + mid_channels, mid_channels, 3, stride, 1, g=mid_channels + ) + self.conv3 = nn.Sequential( + nn.Conv2d(mid_channels, out_channels, 1, bias=False), + nn.BatchNorm2d(out_channels), + ) + + def forward(self, x): + m = self.conv1(x) + m = self.dwconv2(m) + m = self.conv3(m) + if self.use_residual: + return x + m + else: + return m + + +class MobileNetV2(nn.Module): + """MobileNetV2. + + Reference: + Sandler et al. MobileNetV2: Inverted Residuals and + Linear Bottlenecks. CVPR 2018. + + Public keys: + - ``mobilenetv2_x1_0``: MobileNetV2 x1.0. + - ``mobilenetv2_x1_4``: MobileNetV2 x1.4. + """ + + def __init__( + self, + num_classes, + width_mult=1, + loss='softmax', + fc_dims=None, + dropout_p=None, + **kwargs + ): + super(MobileNetV2, self).__init__() + self.loss = loss + self.in_channels = int(32 * width_mult) + self.feature_dim = int(1280 * width_mult) if width_mult > 1 else 1280 + + # construct layers + self.conv1 = ConvBlock(3, self.in_channels, 3, s=2, p=1) + self.conv2 = self._make_layer( + Bottleneck, 1, int(16 * width_mult), 1, 1 + ) + self.conv3 = self._make_layer( + Bottleneck, 6, int(24 * width_mult), 2, 2 + ) + self.conv4 = self._make_layer( + Bottleneck, 6, int(32 * width_mult), 3, 2 + ) + self.conv5 = self._make_layer( + Bottleneck, 6, int(64 * width_mult), 4, 2 + ) + self.conv6 = self._make_layer( + Bottleneck, 6, int(96 * width_mult), 3, 1 + ) + self.conv7 = self._make_layer( + Bottleneck, 6, int(160 * width_mult), 3, 2 + ) + self.conv8 = self._make_layer( + Bottleneck, 6, int(320 * width_mult), 1, 1 + ) + self.conv9 = ConvBlock(self.in_channels, self.feature_dim, 1) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc = self._construct_fc_layer( + fc_dims, self.feature_dim, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _make_layer(self, block, t, c, n, s): + # t: expansion factor + # c: output channels + # n: number of blocks + # s: stride for first layer + layers = [] + layers.append(block(self.in_channels, c, t, s)) + self.in_channels = c + for i in range(1, n): + layers.append(block(self.in_channels, c, t)) + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer. + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.conv2(x) + x = self.conv3(x) + x = self.conv4(x) + x = self.conv5(x) + x = self.conv6(x) + x = self.conv7(x) + x = self.conv8(x) + x = self.conv9(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def mobilenetv2_x1_0(num_classes, loss, pretrained=True, **kwargs): + model = MobileNetV2( + num_classes, + loss=loss, + width_mult=1, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + # init_pretrained_weights(model, model_urls['mobilenetv2_x1_0']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['mobilenetv2_x1_0']) + ) + return model + + +def mobilenetv2_x1_4(num_classes, loss, pretrained=True, **kwargs): + model = MobileNetV2( + num_classes, + loss=loss, + width_mult=1.4, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + # init_pretrained_weights(model, model_urls['mobilenetv2_x1_4']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['mobilenetv2_x1_4']) + ) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py new file mode 100644 index 0000000000..a2ec649d36 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/mudeep.py @@ -0,0 +1,253 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +from torch import nn +from torch.nn import functional as F + +__all__ = ['MuDeep'] + + +class ConvBlock(nn.Module): + """Basic convolutional block. + + convolution + batch normalization + relu. + + Args: + in_c (int): number of input channels. + out_c (int): number of output channels. + k (int or tuple): kernel size. + s (int or tuple): stride. + p (int or tuple): padding. + """ + + def __init__(self, in_c, out_c, k, s, p): + super(ConvBlock, self).__init__() + self.conv = nn.Conv2d(in_c, out_c, k, stride=s, padding=p) + self.bn = nn.BatchNorm2d(out_c) + + def forward(self, x): + return F.relu(self.bn(self.conv(x))) + + +class ConvLayers(nn.Module): + """Preprocessing layers.""" + + def __init__(self): + super(ConvLayers, self).__init__() + self.conv1 = ConvBlock(3, 48, k=3, s=1, p=1) + self.conv2 = ConvBlock(48, 96, k=3, s=1, p=1) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + + def forward(self, x): + x = self.conv1(x) + x = self.conv2(x) + x = self.maxpool(x) + return x + + +class MultiScaleA(nn.Module): + """Multi-scale stream layer A (Sec.3.1)""" + + def __init__(self): + super(MultiScaleA, self).__init__() + self.stream1 = nn.Sequential( + ConvBlock(96, 96, k=1, s=1, p=0), + ConvBlock(96, 24, k=3, s=1, p=1), + ) + self.stream2 = nn.Sequential( + nn.AvgPool2d(kernel_size=3, stride=1, padding=1), + ConvBlock(96, 24, k=1, s=1, p=0), + ) + self.stream3 = ConvBlock(96, 24, k=1, s=1, p=0) + self.stream4 = nn.Sequential( + ConvBlock(96, 16, k=1, s=1, p=0), + ConvBlock(16, 24, k=3, s=1, p=1), + ConvBlock(24, 24, k=3, s=1, p=1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + s4 = self.stream4(x) + y = torch.cat([s1, s2, s3, s4], dim=1) + return y + + +class Reduction(nn.Module): + """Reduction layer (Sec.3.1)""" + + def __init__(self): + super(Reduction, self).__init__() + self.stream1 = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.stream2 = ConvBlock(96, 96, k=3, s=2, p=1) + self.stream3 = nn.Sequential( + ConvBlock(96, 48, k=1, s=1, p=0), + ConvBlock(48, 56, k=3, s=1, p=1), + ConvBlock(56, 64, k=3, s=2, p=1), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + y = torch.cat([s1, s2, s3], dim=1) + return y + + +class MultiScaleB(nn.Module): + """Multi-scale stream layer B (Sec.3.1)""" + + def __init__(self): + super(MultiScaleB, self).__init__() + self.stream1 = nn.Sequential( + nn.AvgPool2d(kernel_size=3, stride=1, padding=1), + ConvBlock(256, 256, k=1, s=1, p=0), + ) + self.stream2 = nn.Sequential( + ConvBlock(256, 64, k=1, s=1, p=0), + ConvBlock(64, 128, k=(1, 3), s=1, p=(0, 1)), + ConvBlock(128, 256, k=(3, 1), s=1, p=(1, 0)), + ) + self.stream3 = ConvBlock(256, 256, k=1, s=1, p=0) + self.stream4 = nn.Sequential( + ConvBlock(256, 64, k=1, s=1, p=0), + ConvBlock(64, 64, k=(1, 3), s=1, p=(0, 1)), + ConvBlock(64, 128, k=(3, 1), s=1, p=(1, 0)), + ConvBlock(128, 128, k=(1, 3), s=1, p=(0, 1)), + ConvBlock(128, 256, k=(3, 1), s=1, p=(1, 0)), + ) + + def forward(self, x): + s1 = self.stream1(x) + s2 = self.stream2(x) + s3 = self.stream3(x) + s4 = self.stream4(x) + return s1, s2, s3, s4 + + +class Fusion(nn.Module): + """Saliency-based learning fusion layer (Sec.3.2)""" + + def __init__(self): + super(Fusion, self).__init__() + self.a1 = nn.Parameter(torch.rand(1, 256, 1, 1)) + self.a2 = nn.Parameter(torch.rand(1, 256, 1, 1)) + self.a3 = nn.Parameter(torch.rand(1, 256, 1, 1)) + self.a4 = nn.Parameter(torch.rand(1, 256, 1, 1)) + + # We add an average pooling layer to reduce the spatial dimension + # of feature maps, which differs from the original paper. + self.avgpool = nn.AvgPool2d(kernel_size=4, stride=4, padding=0) + + def forward(self, x1, x2, x3, x4): + s1 = self.a1.expand_as(x1) * x1 + s2 = self.a2.expand_as(x2) * x2 + s3 = self.a3.expand_as(x3) * x3 + s4 = self.a4.expand_as(x4) * x4 + y = self.avgpool(s1 + s2 + s3 + s4) + return y + + +class MuDeep(nn.Module): + """Multiscale deep neural network. + + Reference: + Qian et al. Multi-scale Deep Learning Architectures + for Person Re-identification. ICCV 2017. + + Public keys: + - ``mudeep``: Multiscale deep neural network. + """ + + def __init__(self, num_classes, loss='softmax', **kwargs): + super(MuDeep, self).__init__() + self.loss = loss + + self.block1 = ConvLayers() + self.block2 = MultiScaleA() + self.block3 = Reduction() + self.block4 = MultiScaleB() + self.block5 = Fusion() + + # Due to this fully connected layer, input image has to be fixed + # in shape, i.e. (3, 256, 128), such that the last convolutional feature + # maps are of shape (256, 16, 8). If input shape is changed, + # the input dimension of this layer has to be changed accordingly. + self.fc = nn.Sequential( + nn.Linear(256 * 16 * 8, 4096), + nn.BatchNorm1d(4096), + nn.ReLU(), + ) + self.classifier = nn.Linear(4096, num_classes) + self.feat_dim = 4096 + + def featuremaps(self, x): + x = self.block1(x) + x = self.block2(x) + x = self.block3(x) + x = self.block4(x) + x = self.block5(*x) + return x + + def forward(self, x): + x = self.featuremaps(x) + x = x.view(x.size(0), -1) + x = self.fc(x) + y = self.classifier(x) + + if not self.training: + return x + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, x + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py new file mode 100644 index 0000000000..dbf951174d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/nasnet.py @@ -0,0 +1,1178 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.nn.functional as F +import torch.utils.model_zoo as model_zoo + +__all__ = ['nasnetamobile'] +""" +NASNet Mobile +Thanks to Anastasiia (https://github.com/DagnyT) for the great help, support and motivation! + + +------------------------------------------------------------------------------------ + Architecture | Top-1 Acc | Top-5 Acc | Multiply-Adds | Params (M) +------------------------------------------------------------------------------------ +| NASNet-A (4 @ 1056) | 74.08% | 91.74% | 564 M | 5.3 | +------------------------------------------------------------------------------------ +# References: + - [Learning Transferable Architectures for Scalable Image Recognition] + (https://arxiv.org/abs/1707.07012) +""" +""" +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" + +pretrained_settings = { + 'nasnetamobile': { + 'imagenet': { + # 'url': 'https://github.com/veronikayurchuk/pretrained-models.pytorch/releases/download/v1.0/nasnetmobile-7e03cead.pth.tar', + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/nasnetamobile-7e03cead.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], # resize 256 + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000 + }, + # 'imagenet+background': { + # # 'url': 'http://data.lip6.fr/cadene/pretrainedmodels/nasnetalarge-a1897284.pth', + # 'input_space': 'RGB', + # 'input_size': [3, 224, 224], # resize 256 + # 'input_range': [0, 1], + # 'mean': [0.5, 0.5, 0.5], + # 'std': [0.5, 0.5, 0.5], + # 'num_classes': 1001 + # } + } +} + + +class MaxPoolPad(nn.Module): + + def __init__(self): + super(MaxPoolPad, self).__init__() + self.pad = nn.ZeroPad2d((1, 0, 1, 0)) + self.pool = nn.MaxPool2d(3, stride=2, padding=1) + + def forward(self, x): + x = self.pad(x) + x = self.pool(x) + x = x[:, :, 1:, 1:].contiguous() + return x + + +class AvgPoolPad(nn.Module): + + def __init__(self, stride=2, padding=1): + super(AvgPoolPad, self).__init__() + self.pad = nn.ZeroPad2d((1, 0, 1, 0)) + self.pool = nn.AvgPool2d( + 3, stride=stride, padding=padding, count_include_pad=False + ) + + def forward(self, x): + x = self.pad(x) + x = self.pool(x) + x = x[:, :, 1:, 1:].contiguous() + return x + + +class SeparableConv2d(nn.Module): + + def __init__( + self, + in_channels, + out_channels, + dw_kernel, + dw_stride, + dw_padding, + bias=False + ): + super(SeparableConv2d, self).__init__() + self.depthwise_conv2d = nn.Conv2d( + in_channels, + in_channels, + dw_kernel, + stride=dw_stride, + padding=dw_padding, + bias=bias, + groups=in_channels + ) + self.pointwise_conv2d = nn.Conv2d( + in_channels, out_channels, 1, stride=1, bias=bias + ) + + def forward(self, x): + x = self.depthwise_conv2d(x) + x = self.pointwise_conv2d(x) + return x + + +class BranchSeparables(nn.Module): + + def __init__( + self, + in_channels, + out_channels, + kernel_size, + stride, + padding, + name=None, + bias=False + ): + super(BranchSeparables, self).__init__() + self.relu = nn.ReLU() + self.separable_1 = SeparableConv2d( + in_channels, in_channels, kernel_size, stride, padding, bias=bias + ) + self.bn_sep_1 = nn.BatchNorm2d( + in_channels, eps=0.001, momentum=0.1, affine=True + ) + self.relu1 = nn.ReLU() + self.separable_2 = SeparableConv2d( + in_channels, out_channels, kernel_size, 1, padding, bias=bias + ) + self.bn_sep_2 = nn.BatchNorm2d( + out_channels, eps=0.001, momentum=0.1, affine=True + ) + self.name = name + + def forward(self, x): + x = self.relu(x) + if self.name == 'specific': + x = nn.ZeroPad2d((1, 0, 1, 0))(x) + x = self.separable_1(x) + if self.name == 'specific': + x = x[:, :, 1:, 1:].contiguous() + + x = self.bn_sep_1(x) + x = self.relu1(x) + x = self.separable_2(x) + x = self.bn_sep_2(x) + return x + + +class BranchSeparablesStem(nn.Module): + + def __init__( + self, + in_channels, + out_channels, + kernel_size, + stride, + padding, + bias=False + ): + super(BranchSeparablesStem, self).__init__() + self.relu = nn.ReLU() + self.separable_1 = SeparableConv2d( + in_channels, out_channels, kernel_size, stride, padding, bias=bias + ) + self.bn_sep_1 = nn.BatchNorm2d( + out_channels, eps=0.001, momentum=0.1, affine=True + ) + self.relu1 = nn.ReLU() + self.separable_2 = SeparableConv2d( + out_channels, out_channels, kernel_size, 1, padding, bias=bias + ) + self.bn_sep_2 = nn.BatchNorm2d( + out_channels, eps=0.001, momentum=0.1, affine=True + ) + + def forward(self, x): + x = self.relu(x) + x = self.separable_1(x) + x = self.bn_sep_1(x) + x = self.relu1(x) + x = self.separable_2(x) + x = self.bn_sep_2(x) + return x + + +class BranchSeparablesReduction(BranchSeparables): + + def __init__( + self, + in_channels, + out_channels, + kernel_size, + stride, + padding, + z_padding=1, + bias=False + ): + BranchSeparables.__init__( + self, in_channels, out_channels, kernel_size, stride, padding, bias + ) + self.padding = nn.ZeroPad2d((z_padding, 0, z_padding, 0)) + + def forward(self, x): + x = self.relu(x) + x = self.padding(x) + x = self.separable_1(x) + x = x[:, :, 1:, 1:].contiguous() + x = self.bn_sep_1(x) + x = self.relu1(x) + x = self.separable_2(x) + x = self.bn_sep_2(x) + return x + + +class CellStem0(nn.Module): + + def __init__(self, stem_filters, num_filters=42): + super(CellStem0, self).__init__() + self.num_filters = num_filters + self.stem_filters = stem_filters + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + self.stem_filters, self.num_filters, 1, stride=1, bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + self.num_filters, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.comb_iter_0_left = BranchSeparables( + self.num_filters, self.num_filters, 5, 2, 2 + ) + self.comb_iter_0_right = BranchSeparablesStem( + self.stem_filters, self.num_filters, 7, 2, 3, bias=False + ) + + self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) + self.comb_iter_1_right = BranchSeparablesStem( + self.stem_filters, self.num_filters, 7, 2, 3, bias=False + ) + + self.comb_iter_2_left = nn.AvgPool2d( + 3, stride=2, padding=1, count_include_pad=False + ) + self.comb_iter_2_right = BranchSeparablesStem( + self.stem_filters, self.num_filters, 5, 2, 2, bias=False + ) + + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparables( + self.num_filters, self.num_filters, 3, 1, 1, bias=False + ) + self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) + + def forward(self, x): + x1 = self.conv_1x1(x) + + x_comb_iter_0_left = self.comb_iter_0_left(x1) + x_comb_iter_0_right = self.comb_iter_0_right(x) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x1) + x_comb_iter_1_right = self.comb_iter_1_right(x) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x1) + x_comb_iter_2_right = self.comb_iter_2_right(x) + x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right + + x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) + x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 + + x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) + x_comb_iter_4_right = self.comb_iter_4_right(x1) + x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right + + x_out = torch.cat( + [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 + ) + return x_out + + +class CellStem1(nn.Module): + + def __init__(self, stem_filters, num_filters): + super(CellStem1, self).__init__() + self.num_filters = num_filters + self.stem_filters = stem_filters + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + 2 * self.num_filters, + self.num_filters, + 1, + stride=1, + bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + self.num_filters, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.relu = nn.ReLU() + self.path_1 = nn.Sequential() + self.path_1.add_module( + 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) + ) + self.path_1.add_module( + 'conv', + nn.Conv2d( + self.stem_filters, + self.num_filters // 2, + 1, + stride=1, + bias=False + ) + ) + self.path_2 = nn.ModuleList() + self.path_2.add_module('pad', nn.ZeroPad2d((0, 1, 0, 1))) + self.path_2.add_module( + 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) + ) + self.path_2.add_module( + 'conv', + nn.Conv2d( + self.stem_filters, + self.num_filters // 2, + 1, + stride=1, + bias=False + ) + ) + + self.final_path_bn = nn.BatchNorm2d( + self.num_filters, eps=0.001, momentum=0.1, affine=True + ) + + self.comb_iter_0_left = BranchSeparables( + self.num_filters, + self.num_filters, + 5, + 2, + 2, + name='specific', + bias=False + ) + self.comb_iter_0_right = BranchSeparables( + self.num_filters, + self.num_filters, + 7, + 2, + 3, + name='specific', + bias=False + ) + + # self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) + self.comb_iter_1_left = MaxPoolPad() + self.comb_iter_1_right = BranchSeparables( + self.num_filters, + self.num_filters, + 7, + 2, + 3, + name='specific', + bias=False + ) + + # self.comb_iter_2_left = nn.AvgPool2d(3, stride=2, padding=1, count_include_pad=False) + self.comb_iter_2_left = AvgPoolPad() + self.comb_iter_2_right = BranchSeparables( + self.num_filters, + self.num_filters, + 5, + 2, + 2, + name='specific', + bias=False + ) + + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparables( + self.num_filters, + self.num_filters, + 3, + 1, + 1, + name='specific', + bias=False + ) + # self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) + self.comb_iter_4_right = MaxPoolPad() + + def forward(self, x_conv0, x_stem_0): + x_left = self.conv_1x1(x_stem_0) + + x_relu = self.relu(x_conv0) + # path 1 + x_path1 = self.path_1(x_relu) + # path 2 + x_path2 = self.path_2.pad(x_relu) + x_path2 = x_path2[:, :, 1:, 1:] + x_path2 = self.path_2.avgpool(x_path2) + x_path2 = self.path_2.conv(x_path2) + # final path + x_right = self.final_path_bn(torch.cat([x_path1, x_path2], 1)) + + x_comb_iter_0_left = self.comb_iter_0_left(x_left) + x_comb_iter_0_right = self.comb_iter_0_right(x_right) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x_left) + x_comb_iter_1_right = self.comb_iter_1_right(x_right) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x_left) + x_comb_iter_2_right = self.comb_iter_2_right(x_right) + x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right + + x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) + x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 + + x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) + x_comb_iter_4_right = self.comb_iter_4_right(x_left) + x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right + + x_out = torch.cat( + [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 + ) + return x_out + + +class FirstCell(nn.Module): + + def __init__( + self, in_channels_left, out_channels_left, in_channels_right, + out_channels_right + ): + super(FirstCell, self).__init__() + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_right, out_channels_right, 1, stride=1, bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_right, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.relu = nn.ReLU() + self.path_1 = nn.Sequential() + self.path_1.add_module( + 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) + ) + self.path_1.add_module( + 'conv', + nn.Conv2d( + in_channels_left, out_channels_left, 1, stride=1, bias=False + ) + ) + self.path_2 = nn.ModuleList() + self.path_2.add_module('pad', nn.ZeroPad2d((0, 1, 0, 1))) + self.path_2.add_module( + 'avgpool', nn.AvgPool2d(1, stride=2, count_include_pad=False) + ) + self.path_2.add_module( + 'conv', + nn.Conv2d( + in_channels_left, out_channels_left, 1, stride=1, bias=False + ) + ) + + self.final_path_bn = nn.BatchNorm2d( + out_channels_left * 2, eps=0.001, momentum=0.1, affine=True + ) + + self.comb_iter_0_left = BranchSeparables( + out_channels_right, out_channels_right, 5, 1, 2, bias=False + ) + self.comb_iter_0_right = BranchSeparables( + out_channels_right, out_channels_right, 3, 1, 1, bias=False + ) + + self.comb_iter_1_left = BranchSeparables( + out_channels_right, out_channels_right, 5, 1, 2, bias=False + ) + self.comb_iter_1_right = BranchSeparables( + out_channels_right, out_channels_right, 3, 1, 1, bias=False + ) + + self.comb_iter_2_left = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_3_left = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparables( + out_channels_right, out_channels_right, 3, 1, 1, bias=False + ) + + def forward(self, x, x_prev): + x_relu = self.relu(x_prev) + # path 1 + x_path1 = self.path_1(x_relu) + # path 2 + x_path2 = self.path_2.pad(x_relu) + x_path2 = x_path2[:, :, 1:, 1:] + x_path2 = self.path_2.avgpool(x_path2) + x_path2 = self.path_2.conv(x_path2) + # final path + x_left = self.final_path_bn(torch.cat([x_path1, x_path2], 1)) + + x_right = self.conv_1x1(x) + + x_comb_iter_0_left = self.comb_iter_0_left(x_right) + x_comb_iter_0_right = self.comb_iter_0_right(x_left) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x_left) + x_comb_iter_1_right = self.comb_iter_1_right(x_left) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x_right) + x_comb_iter_2 = x_comb_iter_2_left + x_left + + x_comb_iter_3_left = self.comb_iter_3_left(x_left) + x_comb_iter_3_right = self.comb_iter_3_right(x_left) + x_comb_iter_3 = x_comb_iter_3_left + x_comb_iter_3_right + + x_comb_iter_4_left = self.comb_iter_4_left(x_right) + x_comb_iter_4 = x_comb_iter_4_left + x_right + + x_out = torch.cat( + [ + x_left, x_comb_iter_0, x_comb_iter_1, x_comb_iter_2, + x_comb_iter_3, x_comb_iter_4 + ], 1 + ) + return x_out + + +class NormalCell(nn.Module): + + def __init__( + self, in_channels_left, out_channels_left, in_channels_right, + out_channels_right + ): + super(NormalCell, self).__init__() + self.conv_prev_1x1 = nn.Sequential() + self.conv_prev_1x1.add_module('relu', nn.ReLU()) + self.conv_prev_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_left, out_channels_left, 1, stride=1, bias=False + ) + ) + self.conv_prev_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_left, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_right, out_channels_right, 1, stride=1, bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_right, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.comb_iter_0_left = BranchSeparables( + out_channels_right, out_channels_right, 5, 1, 2, bias=False + ) + self.comb_iter_0_right = BranchSeparables( + out_channels_left, out_channels_left, 3, 1, 1, bias=False + ) + + self.comb_iter_1_left = BranchSeparables( + out_channels_left, out_channels_left, 5, 1, 2, bias=False + ) + self.comb_iter_1_right = BranchSeparables( + out_channels_left, out_channels_left, 3, 1, 1, bias=False + ) + + self.comb_iter_2_left = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_3_left = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparables( + out_channels_right, out_channels_right, 3, 1, 1, bias=False + ) + + def forward(self, x, x_prev): + x_left = self.conv_prev_1x1(x_prev) + x_right = self.conv_1x1(x) + + x_comb_iter_0_left = self.comb_iter_0_left(x_right) + x_comb_iter_0_right = self.comb_iter_0_right(x_left) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x_left) + x_comb_iter_1_right = self.comb_iter_1_right(x_left) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x_right) + x_comb_iter_2 = x_comb_iter_2_left + x_left + + x_comb_iter_3_left = self.comb_iter_3_left(x_left) + x_comb_iter_3_right = self.comb_iter_3_right(x_left) + x_comb_iter_3 = x_comb_iter_3_left + x_comb_iter_3_right + + x_comb_iter_4_left = self.comb_iter_4_left(x_right) + x_comb_iter_4 = x_comb_iter_4_left + x_right + + x_out = torch.cat( + [ + x_left, x_comb_iter_0, x_comb_iter_1, x_comb_iter_2, + x_comb_iter_3, x_comb_iter_4 + ], 1 + ) + return x_out + + +class ReductionCell0(nn.Module): + + def __init__( + self, in_channels_left, out_channels_left, in_channels_right, + out_channels_right + ): + super(ReductionCell0, self).__init__() + self.conv_prev_1x1 = nn.Sequential() + self.conv_prev_1x1.add_module('relu', nn.ReLU()) + self.conv_prev_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_left, out_channels_left, 1, stride=1, bias=False + ) + ) + self.conv_prev_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_left, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_right, out_channels_right, 1, stride=1, bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_right, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.comb_iter_0_left = BranchSeparablesReduction( + out_channels_right, out_channels_right, 5, 2, 2, bias=False + ) + self.comb_iter_0_right = BranchSeparablesReduction( + out_channels_right, out_channels_right, 7, 2, 3, bias=False + ) + + self.comb_iter_1_left = MaxPoolPad() + self.comb_iter_1_right = BranchSeparablesReduction( + out_channels_right, out_channels_right, 7, 2, 3, bias=False + ) + + self.comb_iter_2_left = AvgPoolPad() + self.comb_iter_2_right = BranchSeparablesReduction( + out_channels_right, out_channels_right, 5, 2, 2, bias=False + ) + + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparablesReduction( + out_channels_right, out_channels_right, 3, 1, 1, bias=False + ) + self.comb_iter_4_right = MaxPoolPad() + + def forward(self, x, x_prev): + x_left = self.conv_prev_1x1(x_prev) + x_right = self.conv_1x1(x) + + x_comb_iter_0_left = self.comb_iter_0_left(x_right) + x_comb_iter_0_right = self.comb_iter_0_right(x_left) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x_right) + x_comb_iter_1_right = self.comb_iter_1_right(x_left) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x_right) + x_comb_iter_2_right = self.comb_iter_2_right(x_left) + x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right + + x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) + x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 + + x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) + x_comb_iter_4_right = self.comb_iter_4_right(x_right) + x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right + + x_out = torch.cat( + [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 + ) + return x_out + + +class ReductionCell1(nn.Module): + + def __init__( + self, in_channels_left, out_channels_left, in_channels_right, + out_channels_right + ): + super(ReductionCell1, self).__init__() + self.conv_prev_1x1 = nn.Sequential() + self.conv_prev_1x1.add_module('relu', nn.ReLU()) + self.conv_prev_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_left, out_channels_left, 1, stride=1, bias=False + ) + ) + self.conv_prev_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_left, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.conv_1x1 = nn.Sequential() + self.conv_1x1.add_module('relu', nn.ReLU()) + self.conv_1x1.add_module( + 'conv', + nn.Conv2d( + in_channels_right, out_channels_right, 1, stride=1, bias=False + ) + ) + self.conv_1x1.add_module( + 'bn', + nn.BatchNorm2d( + out_channels_right, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.comb_iter_0_left = BranchSeparables( + out_channels_right, + out_channels_right, + 5, + 2, + 2, + name='specific', + bias=False + ) + self.comb_iter_0_right = BranchSeparables( + out_channels_right, + out_channels_right, + 7, + 2, + 3, + name='specific', + bias=False + ) + + # self.comb_iter_1_left = nn.MaxPool2d(3, stride=2, padding=1) + self.comb_iter_1_left = MaxPoolPad() + self.comb_iter_1_right = BranchSeparables( + out_channels_right, + out_channels_right, + 7, + 2, + 3, + name='specific', + bias=False + ) + + # self.comb_iter_2_left = nn.AvgPool2d(3, stride=2, padding=1, count_include_pad=False) + self.comb_iter_2_left = AvgPoolPad() + self.comb_iter_2_right = BranchSeparables( + out_channels_right, + out_channels_right, + 5, + 2, + 2, + name='specific', + bias=False + ) + + self.comb_iter_3_right = nn.AvgPool2d( + 3, stride=1, padding=1, count_include_pad=False + ) + + self.comb_iter_4_left = BranchSeparables( + out_channels_right, + out_channels_right, + 3, + 1, + 1, + name='specific', + bias=False + ) + # self.comb_iter_4_right = nn.MaxPool2d(3, stride=2, padding=1) + self.comb_iter_4_right = MaxPoolPad() + + def forward(self, x, x_prev): + x_left = self.conv_prev_1x1(x_prev) + x_right = self.conv_1x1(x) + + x_comb_iter_0_left = self.comb_iter_0_left(x_right) + x_comb_iter_0_right = self.comb_iter_0_right(x_left) + x_comb_iter_0 = x_comb_iter_0_left + x_comb_iter_0_right + + x_comb_iter_1_left = self.comb_iter_1_left(x_right) + x_comb_iter_1_right = self.comb_iter_1_right(x_left) + x_comb_iter_1 = x_comb_iter_1_left + x_comb_iter_1_right + + x_comb_iter_2_left = self.comb_iter_2_left(x_right) + x_comb_iter_2_right = self.comb_iter_2_right(x_left) + x_comb_iter_2 = x_comb_iter_2_left + x_comb_iter_2_right + + x_comb_iter_3_right = self.comb_iter_3_right(x_comb_iter_0) + x_comb_iter_3 = x_comb_iter_3_right + x_comb_iter_1 + + x_comb_iter_4_left = self.comb_iter_4_left(x_comb_iter_0) + x_comb_iter_4_right = self.comb_iter_4_right(x_right) + x_comb_iter_4 = x_comb_iter_4_left + x_comb_iter_4_right + + x_out = torch.cat( + [x_comb_iter_1, x_comb_iter_2, x_comb_iter_3, x_comb_iter_4], 1 + ) + return x_out + + +class NASNetAMobile(nn.Module): + """Neural Architecture Search (NAS). + + Reference: + Zoph et al. Learning Transferable Architectures + for Scalable Image Recognition. CVPR 2018. + + Public keys: + - ``nasnetamobile``: NASNet-A Mobile. + """ + + def __init__( + self, + num_classes, + loss, + stem_filters=32, + penultimate_filters=1056, + filters_multiplier=2, + **kwargs + ): + super(NASNetAMobile, self).__init__() + self.stem_filters = stem_filters + self.penultimate_filters = penultimate_filters + self.filters_multiplier = filters_multiplier + self.loss = loss + + filters = self.penultimate_filters // 24 + # 24 is default value for the architecture + + self.conv0 = nn.Sequential() + self.conv0.add_module( + 'conv', + nn.Conv2d( + in_channels=3, + out_channels=self.stem_filters, + kernel_size=3, + padding=0, + stride=2, + bias=False + ) + ) + self.conv0.add_module( + 'bn', + nn.BatchNorm2d( + self.stem_filters, eps=0.001, momentum=0.1, affine=True + ) + ) + + self.cell_stem_0 = CellStem0( + self.stem_filters, num_filters=filters // (filters_multiplier**2) + ) + self.cell_stem_1 = CellStem1( + self.stem_filters, num_filters=filters // filters_multiplier + ) + + self.cell_0 = FirstCell( + in_channels_left=filters, + out_channels_left=filters // 2, # 1, 0.5 + in_channels_right=2 * filters, + out_channels_right=filters + ) # 2, 1 + self.cell_1 = NormalCell( + in_channels_left=2 * filters, + out_channels_left=filters, # 2, 1 + in_channels_right=6 * filters, + out_channels_right=filters + ) # 6, 1 + self.cell_2 = NormalCell( + in_channels_left=6 * filters, + out_channels_left=filters, # 6, 1 + in_channels_right=6 * filters, + out_channels_right=filters + ) # 6, 1 + self.cell_3 = NormalCell( + in_channels_left=6 * filters, + out_channels_left=filters, # 6, 1 + in_channels_right=6 * filters, + out_channels_right=filters + ) # 6, 1 + + self.reduction_cell_0 = ReductionCell0( + in_channels_left=6 * filters, + out_channels_left=2 * filters, # 6, 2 + in_channels_right=6 * filters, + out_channels_right=2 * filters + ) # 6, 2 + + self.cell_6 = FirstCell( + in_channels_left=6 * filters, + out_channels_left=filters, # 6, 1 + in_channels_right=8 * filters, + out_channels_right=2 * filters + ) # 8, 2 + self.cell_7 = NormalCell( + in_channels_left=8 * filters, + out_channels_left=2 * filters, # 8, 2 + in_channels_right=12 * filters, + out_channels_right=2 * filters + ) # 12, 2 + self.cell_8 = NormalCell( + in_channels_left=12 * filters, + out_channels_left=2 * filters, # 12, 2 + in_channels_right=12 * filters, + out_channels_right=2 * filters + ) # 12, 2 + self.cell_9 = NormalCell( + in_channels_left=12 * filters, + out_channels_left=2 * filters, # 12, 2 + in_channels_right=12 * filters, + out_channels_right=2 * filters + ) # 12, 2 + + self.reduction_cell_1 = ReductionCell1( + in_channels_left=12 * filters, + out_channels_left=4 * filters, # 12, 4 + in_channels_right=12 * filters, + out_channels_right=4 * filters + ) # 12, 4 + + self.cell_12 = FirstCell( + in_channels_left=12 * filters, + out_channels_left=2 * filters, # 12, 2 + in_channels_right=16 * filters, + out_channels_right=4 * filters + ) # 16, 4 + self.cell_13 = NormalCell( + in_channels_left=16 * filters, + out_channels_left=4 * filters, # 16, 4 + in_channels_right=24 * filters, + out_channels_right=4 * filters + ) # 24, 4 + self.cell_14 = NormalCell( + in_channels_left=24 * filters, + out_channels_left=4 * filters, # 24, 4 + in_channels_right=24 * filters, + out_channels_right=4 * filters + ) # 24, 4 + self.cell_15 = NormalCell( + in_channels_left=24 * filters, + out_channels_left=4 * filters, # 24, 4 + in_channels_right=24 * filters, + out_channels_right=4 * filters + ) # 24, 4 + + self.relu = nn.ReLU() + self.dropout = nn.Dropout() + self.classifier = nn.Linear(24 * filters, num_classes) + + self._init_params() + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def features(self, input): + x_conv0 = self.conv0(input) + x_stem_0 = self.cell_stem_0(x_conv0) + x_stem_1 = self.cell_stem_1(x_conv0, x_stem_0) + + x_cell_0 = self.cell_0(x_stem_1, x_stem_0) + x_cell_1 = self.cell_1(x_cell_0, x_stem_1) + x_cell_2 = self.cell_2(x_cell_1, x_cell_0) + x_cell_3 = self.cell_3(x_cell_2, x_cell_1) + + x_reduction_cell_0 = self.reduction_cell_0(x_cell_3, x_cell_2) + + x_cell_6 = self.cell_6(x_reduction_cell_0, x_cell_3) + x_cell_7 = self.cell_7(x_cell_6, x_reduction_cell_0) + x_cell_8 = self.cell_8(x_cell_7, x_cell_6) + x_cell_9 = self.cell_9(x_cell_8, x_cell_7) + + x_reduction_cell_1 = self.reduction_cell_1(x_cell_9, x_cell_8) + + x_cell_12 = self.cell_12(x_reduction_cell_1, x_cell_9) + x_cell_13 = self.cell_13(x_cell_12, x_reduction_cell_1) + x_cell_14 = self.cell_14(x_cell_13, x_cell_12) + x_cell_15 = self.cell_15(x_cell_14, x_cell_13) + + x_cell_15 = self.relu(x_cell_15) + x_cell_15 = F.avg_pool2d( + x_cell_15, + x_cell_15.size()[2:] + ) # global average pool + x_cell_15 = x_cell_15.view(x_cell_15.size(0), -1) + x_cell_15 = self.dropout(x_cell_15) + + return x_cell_15 + + def forward(self, input): + v = self.features(input) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def nasnetamobile(num_classes, loss='softmax', pretrained=True, **kwargs): + model = NASNetAMobile(num_classes, loss, **kwargs) + if pretrained: + model_url = pretrained_settings['nasnetamobile']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py new file mode 100644 index 0000000000..f831d02648 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet.py @@ -0,0 +1,645 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import warnings +import torch +from torch import nn +from torch.nn import functional as F + +__all__ = [ + 'osnet_x1_0', 'osnet_x0_75', 'osnet_x0_5', 'osnet_x0_25', 'osnet_ibn_x1_0' +] + +pretrained_urls = { + 'osnet_x1_0': + 'https://drive.google.com/uc?id=1LaG1EJpHrxdAxKnSCJ_i0u-nbxSAeiFY', + 'osnet_x0_75': + 'https://drive.google.com/uc?id=1uwA9fElHOk3ZogwbeY5GkLI6QPTX70Hq', + 'osnet_x0_5': + 'https://drive.google.com/uc?id=16DGLbZukvVYgINws8u8deSaOqjybZ83i', + 'osnet_x0_25': + 'https://drive.google.com/uc?id=1rb8UN5ZzPKRc_xvtHlyDh-cSz88YX9hs', + 'osnet_ibn_x1_0': + 'https://drive.google.com/uc?id=1sr90V6irlYYDd4_4ISU2iruoRG8J__6l' +} + + +########## +# Basic layers +########## +class ConvLayer(nn.Module): + """Convolution layer (conv + bn + relu).""" + + def __init__( + self, + in_channels, + out_channels, + kernel_size, + stride=1, + padding=0, + groups=1, + IN=False + ): + super(ConvLayer, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + kernel_size, + stride=stride, + padding=padding, + bias=False, + groups=groups + ) + if IN: + self.bn = nn.InstanceNorm2d(out_channels, affine=True) + else: + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Conv1x1(nn.Module): + """1x1 convolution + bn + relu.""" + + def __init__(self, in_channels, out_channels, stride=1, groups=1): + super(Conv1x1, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + 1, + stride=stride, + padding=0, + bias=False, + groups=groups + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class Conv1x1Linear(nn.Module): + """1x1 convolution + bn (w/o non-linearity).""" + + def __init__(self, in_channels, out_channels, stride=1): + super(Conv1x1Linear, self).__init__() + self.conv = nn.Conv2d( + in_channels, out_channels, 1, stride=stride, padding=0, bias=False + ) + self.bn = nn.BatchNorm2d(out_channels) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + return x + + +class Conv3x3(nn.Module): + """3x3 convolution + bn + relu.""" + + def __init__(self, in_channels, out_channels, stride=1, groups=1): + super(Conv3x3, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + 3, + stride=stride, + padding=1, + bias=False, + groups=groups + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + x = self.relu(x) + return x + + +class LightConv3x3(nn.Module): + """Lightweight 3x3 convolution. + + 1x1 (linear) + dw 3x3 (nonlinear). + """ + + def __init__(self, in_channels, out_channels): + super(LightConv3x3, self).__init__() + self.conv1 = nn.Conv2d( + in_channels, out_channels, 1, stride=1, padding=0, bias=False + ) + self.conv2 = nn.Conv2d( + out_channels, + out_channels, + 3, + stride=1, + padding=1, + bias=False, + groups=out_channels + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.conv1(x) + x = self.conv2(x) + x = self.bn(x) + x = self.relu(x) + return x + + +########## +# Building blocks for omni-scale feature learning +########## +class ChannelGate(nn.Module): + """A mini-network that generates channel-wise gates conditioned on input tensor.""" + + def __init__( + self, + in_channels, + num_gates=None, + return_gates=False, + gate_activation='sigmoid', + reduction=16, + layer_norm=False + ): + super(ChannelGate, self).__init__() + if num_gates is None: + num_gates = in_channels + self.return_gates = return_gates + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc1 = nn.Conv2d( + in_channels, + in_channels // reduction, + kernel_size=1, + bias=True, + padding=0 + ) + self.norm1 = None + if layer_norm: + self.norm1 = nn.LayerNorm((in_channels // reduction, 1, 1)) + self.relu = nn.ReLU(inplace=True) + self.fc2 = nn.Conv2d( + in_channels // reduction, + num_gates, + kernel_size=1, + bias=True, + padding=0 + ) + if gate_activation == 'sigmoid': + self.gate_activation = nn.Sigmoid() + elif gate_activation == 'relu': + self.gate_activation = nn.ReLU(inplace=True) + elif gate_activation == 'linear': + self.gate_activation = None + else: + raise RuntimeError( + "Unknown gate activation: {}".format(gate_activation) + ) + + def forward(self, x): + input = x + x = self.global_avgpool(x) + x = self.fc1(x) + if self.norm1 is not None: + x = self.norm1(x) + x = self.relu(x) + x = self.fc2(x) + if self.gate_activation is not None: + x = self.gate_activation(x) + if self.return_gates: + return x + return input * x + + +class OSBlock(nn.Module): + """Omni-scale feature learning block.""" + + def __init__( + self, + in_channels, + out_channels, + IN=False, + bottleneck_reduction=4, + **kwargs + ): + super(OSBlock, self).__init__() + mid_channels = out_channels // bottleneck_reduction + self.conv1 = Conv1x1(in_channels, mid_channels) + self.conv2a = LightConv3x3(mid_channels, mid_channels) + self.conv2b = nn.Sequential( + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + ) + self.conv2c = nn.Sequential( + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + ) + self.conv2d = nn.Sequential( + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + LightConv3x3(mid_channels, mid_channels), + ) + self.gate = ChannelGate(mid_channels) + self.conv3 = Conv1x1Linear(mid_channels, out_channels) + self.downsample = None + if in_channels != out_channels: + self.downsample = Conv1x1Linear(in_channels, out_channels) + self.IN = None + if IN: + self.IN = nn.InstanceNorm2d(out_channels, affine=True) + + def forward(self, x): + identity = x + x1 = self.conv1(x) + x2a = self.conv2a(x1) + x2b = self.conv2b(x1) + x2c = self.conv2c(x1) + x2d = self.conv2d(x1) + x2 = self.gate(x2a) + self.gate(x2b) + self.gate(x2c) + self.gate(x2d) + x3 = self.conv3(x2) + if self.downsample is not None: + identity = self.downsample(identity) + out = x3 + identity + if self.IN is not None: + out = self.IN(out) + return F.relu(out) + + +########## +# Network architecture +########## +class OSNet(nn.Module): + """Omni-Scale Network. + + Reference: + - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. + - Zhou et al. Learning Generalisable Omni-Scale Representations + for Person Re-Identification. TPAMI, 2021. + """ + + def __init__( + self, + num_classes, + blocks, + layers, + channels, + feature_dim=512, + loss='softmax', + IN=False, + **kwargs + ): + super(OSNet, self).__init__() + num_blocks = len(blocks) + assert num_blocks == len(layers) + assert num_blocks == len(channels) - 1 + self.loss = loss + self.feature_dim = feature_dim + + # convolutional backbone + self.conv1 = ConvLayer(3, channels[0], 7, stride=2, padding=3, IN=IN) + self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) + self.conv2 = self._make_layer( + blocks[0], + layers[0], + channels[0], + channels[1], + reduce_spatial_size=True, + IN=IN + ) + self.conv3 = self._make_layer( + blocks[1], + layers[1], + channels[1], + channels[2], + reduce_spatial_size=True + ) + self.conv4 = self._make_layer( + blocks[2], + layers[2], + channels[2], + channels[3], + reduce_spatial_size=False + ) + self.conv5 = Conv1x1(channels[3], channels[3]) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + # fully connected layer + self.fc = self._construct_fc_layer( + self.feature_dim, channels[3], dropout_p=None + ) + # identity classification layer + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _make_layer( + self, + block, + layer, + in_channels, + out_channels, + reduce_spatial_size, + IN=False + ): + layers = [] + + layers.append(block(in_channels, out_channels, IN=IN)) + for i in range(1, layer): + layers.append(block(out_channels, out_channels, IN=IN)) + + if reduce_spatial_size: + layers.append( + nn.Sequential( + Conv1x1(out_channels, out_channels), + nn.AvgPool2d(2, stride=2) + ) + ) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + if fc_dims is None or fc_dims < 0: + self.feature_dim = input_dim + return None + + if isinstance(fc_dims, int): + fc_dims = [fc_dims] + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.maxpool(x) + x = self.conv2(x) + x = self.conv3(x) + x = self.conv4(x) + x = self.conv5(x) + return x + + def forward(self, x, return_featuremaps=False): + x = self.featuremaps(x) + if return_featuremaps: + return x + v = self.global_avgpool(x) + v = v.view(v.size(0), -1) + if self.fc is not None: + v = self.fc(v) + if not self.training: + return v + y = self.classifier(v) + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, key=''): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + import os + import errno + import gdown + from collections import OrderedDict + + def _get_torch_home(): + ENV_TORCH_HOME = 'TORCH_HOME' + ENV_XDG_CACHE_HOME = 'XDG_CACHE_HOME' + DEFAULT_CACHE_DIR = '~/.cache' + torch_home = os.path.expanduser( + os.getenv( + ENV_TORCH_HOME, + os.path.join( + os.getenv(ENV_XDG_CACHE_HOME, DEFAULT_CACHE_DIR), 'torch' + ) + ) + ) + return torch_home + + torch_home = _get_torch_home() + model_dir = os.path.join(torch_home, 'checkpoints') + try: + os.makedirs(model_dir) + except OSError as e: + if e.errno == errno.EEXIST: + # Directory already exists, ignore. + pass + else: + # Unexpected OSError, re-raise. + raise + filename = key + '_imagenet.pth' + cached_file = os.path.join(model_dir, filename) + + if not os.path.exists(cached_file): + gdown.download(pretrained_urls[key], cached_file, quiet=False) + + state_dict = torch.load(cached_file) + model_dict = model.state_dict() + new_state_dict = OrderedDict() + matched_layers, discarded_layers = [], [] + + for k, v in state_dict.items(): + if k.startswith('module.'): + k = k[7:] # discard module. + + if k in model_dict and model_dict[k].size() == v.size(): + new_state_dict[k] = v + matched_layers.append(k) + else: + discarded_layers.append(k) + + model_dict.update(new_state_dict) + model.load_state_dict(model_dict) + + if len(matched_layers) == 0: + warnings.warn( + 'The pretrained weights from "{}" cannot be loaded, ' + 'please check the key names manually ' + '(** ignored and continue **)'.format(cached_file) + ) + else: + print( + 'Successfully loaded imagenet pretrained weights from "{}"'. + format(cached_file) + ) + if len(discarded_layers) > 0: + print( + '** The following layers are discarded ' + 'due to unmatched keys or layer size: {}'. + format(discarded_layers) + ) + + +########## +# Instantiation +########## +def osnet_x1_0(num_classes=1000, pretrained=True, loss='softmax', **kwargs): + # standard size (width x1.0) + model = OSNet( + num_classes, + blocks=[OSBlock, OSBlock, OSBlock], + layers=[2, 2, 2], + channels=[64, 256, 384, 512], + loss=loss, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_x1_0') + return model + + +def osnet_x0_75(num_classes=1000, pretrained=True, loss='softmax', **kwargs): + # medium size (width x0.75) + model = OSNet( + num_classes, + blocks=[OSBlock, OSBlock, OSBlock], + layers=[2, 2, 2], + channels=[48, 192, 288, 384], + loss=loss, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_x0_75') + return model + + +def osnet_x0_5(num_classes=1000, pretrained=True, loss='softmax', **kwargs): + # tiny size (width x0.5) + model = OSNet( + num_classes, + blocks=[OSBlock, OSBlock, OSBlock], + layers=[2, 2, 2], + channels=[32, 128, 192, 256], + loss=loss, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_x0_5') + return model + + +def osnet_x0_25(num_classes=1000, pretrained=True, loss='softmax', **kwargs): + # very tiny size (width x0.25) + model = OSNet( + num_classes, + blocks=[OSBlock, OSBlock, OSBlock], + layers=[2, 2, 2], + channels=[16, 64, 96, 128], + loss=loss, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_x0_25') + return model + + +def osnet_ibn_x1_0( + num_classes=1000, pretrained=True, loss='softmax', **kwargs +): + # standard size (width x1.0) + IBN layer + # Ref: Pan et al. Two at Once: Enhancing Learning and Generalization Capacities via IBN-Net. ECCV, 2018. + model = OSNet( + num_classes, + blocks=[OSBlock, OSBlock, OSBlock], + layers=[2, 2, 2], + channels=[64, 256, 384, 512], + loss=loss, + IN=True, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_ibn_x1_0') + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py new file mode 100644 index 0000000000..3f71e7a528 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/osnet_ain.py @@ -0,0 +1,588 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import warnings +import torch +from torch import nn +from torch.nn import functional as F + +__all__ = ['osnet_ain_x1_0'] + +pretrained_urls = { + 'osnet_ain_x1_0': + 'https://drive.google.com/uc?id=1-CaioD9NaqbHK_kzSMW8VE4_3KcsRjEo' +} + + +########## +# Basic layers +########## +class ConvLayer(nn.Module): + """Convolution layer (conv + bn + relu).""" + + def __init__( + self, + in_channels, + out_channels, + kernel_size, + stride=1, + padding=0, + groups=1, + IN=False + ): + super(ConvLayer, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + kernel_size, + stride=stride, + padding=padding, + bias=False, + groups=groups + ) + if IN: + self.bn = nn.InstanceNorm2d(out_channels, affine=True) + else: + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU() + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + return self.relu(x) + + +class Conv1x1(nn.Module): + """1x1 convolution + bn + relu.""" + + def __init__(self, in_channels, out_channels, stride=1, groups=1): + super(Conv1x1, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + 1, + stride=stride, + padding=0, + bias=False, + groups=groups + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU() + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + return self.relu(x) + + +class Conv1x1Linear(nn.Module): + """1x1 convolution + bn (w/o non-linearity).""" + + def __init__(self, in_channels, out_channels, stride=1, bn=True): + super(Conv1x1Linear, self).__init__() + self.conv = nn.Conv2d( + in_channels, out_channels, 1, stride=stride, padding=0, bias=False + ) + self.bn = None + if bn: + self.bn = nn.BatchNorm2d(out_channels) + + def forward(self, x): + x = self.conv(x) + if self.bn is not None: + x = self.bn(x) + return x + + +class Conv3x3(nn.Module): + """3x3 convolution + bn + relu.""" + + def __init__(self, in_channels, out_channels, stride=1, groups=1): + super(Conv3x3, self).__init__() + self.conv = nn.Conv2d( + in_channels, + out_channels, + 3, + stride=stride, + padding=1, + bias=False, + groups=groups + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU() + + def forward(self, x): + x = self.conv(x) + x = self.bn(x) + return self.relu(x) + + +class LightConv3x3(nn.Module): + """Lightweight 3x3 convolution. + + 1x1 (linear) + dw 3x3 (nonlinear). + """ + + def __init__(self, in_channels, out_channels): + super(LightConv3x3, self).__init__() + self.conv1 = nn.Conv2d( + in_channels, out_channels, 1, stride=1, padding=0, bias=False + ) + self.conv2 = nn.Conv2d( + out_channels, + out_channels, + 3, + stride=1, + padding=1, + bias=False, + groups=out_channels + ) + self.bn = nn.BatchNorm2d(out_channels) + self.relu = nn.ReLU() + + def forward(self, x): + x = self.conv1(x) + x = self.conv2(x) + x = self.bn(x) + return self.relu(x) + + +class LightConvStream(nn.Module): + """Lightweight convolution stream.""" + + def __init__(self, in_channels, out_channels, depth): + super(LightConvStream, self).__init__() + assert depth >= 1, 'depth must be equal to or larger than 1, but got {}'.format( + depth + ) + layers = [] + layers += [LightConv3x3(in_channels, out_channels)] + for i in range(depth - 1): + layers += [LightConv3x3(out_channels, out_channels)] + self.layers = nn.Sequential(*layers) + + def forward(self, x): + return self.layers(x) + + +########## +# Building blocks for omni-scale feature learning +########## +class ChannelGate(nn.Module): + """A mini-network that generates channel-wise gates conditioned on input tensor.""" + + def __init__( + self, + in_channels, + num_gates=None, + return_gates=False, + gate_activation='sigmoid', + reduction=16, + layer_norm=False + ): + super(ChannelGate, self).__init__() + if num_gates is None: + num_gates = in_channels + self.return_gates = return_gates + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc1 = nn.Conv2d( + in_channels, + in_channels // reduction, + kernel_size=1, + bias=True, + padding=0 + ) + self.norm1 = None + if layer_norm: + self.norm1 = nn.LayerNorm((in_channels // reduction, 1, 1)) + self.relu = nn.ReLU() + self.fc2 = nn.Conv2d( + in_channels // reduction, + num_gates, + kernel_size=1, + bias=True, + padding=0 + ) + if gate_activation == 'sigmoid': + self.gate_activation = nn.Sigmoid() + elif gate_activation == 'relu': + self.gate_activation = nn.ReLU() + elif gate_activation == 'linear': + self.gate_activation = None + else: + raise RuntimeError( + "Unknown gate activation: {}".format(gate_activation) + ) + + def forward(self, x): + input = x + x = self.global_avgpool(x) + x = self.fc1(x) + if self.norm1 is not None: + x = self.norm1(x) + x = self.relu(x) + x = self.fc2(x) + if self.gate_activation is not None: + x = self.gate_activation(x) + if self.return_gates: + return x + return input * x + + +class OSBlock(nn.Module): + """Omni-scale feature learning block.""" + + def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs): + super(OSBlock, self).__init__() + assert T >= 1 + assert out_channels >= reduction and out_channels % reduction == 0 + mid_channels = out_channels // reduction + + self.conv1 = Conv1x1(in_channels, mid_channels) + self.conv2 = nn.ModuleList() + for t in range(1, T + 1): + self.conv2 += [LightConvStream(mid_channels, mid_channels, t)] + self.gate = ChannelGate(mid_channels) + self.conv3 = Conv1x1Linear(mid_channels, out_channels) + self.downsample = None + if in_channels != out_channels: + self.downsample = Conv1x1Linear(in_channels, out_channels) + + def forward(self, x): + identity = x + x1 = self.conv1(x) + x2 = 0 + for conv2_t in self.conv2: + x2_t = conv2_t(x1) + x2 = x2 + self.gate(x2_t) + x3 = self.conv3(x2) + if self.downsample is not None: + identity = self.downsample(identity) + out = x3 + identity + return F.relu(out) + + +class OSBlockINin(nn.Module): + """Omni-scale feature learning block with instance normalization.""" + + def __init__(self, in_channels, out_channels, reduction=4, T=4, **kwargs): + super(OSBlockINin, self).__init__() + assert T >= 1 + assert out_channels >= reduction and out_channels % reduction == 0 + mid_channels = out_channels // reduction + + self.conv1 = Conv1x1(in_channels, mid_channels) + self.conv2 = nn.ModuleList() + for t in range(1, T + 1): + self.conv2 += [LightConvStream(mid_channels, mid_channels, t)] + self.gate = ChannelGate(mid_channels) + self.conv3 = Conv1x1Linear(mid_channels, out_channels, bn=False) + self.downsample = None + if in_channels != out_channels: + self.downsample = Conv1x1Linear(in_channels, out_channels) + self.IN = nn.InstanceNorm2d(out_channels, affine=True) + + def forward(self, x): + identity = x + x1 = self.conv1(x) + x2 = 0 + for conv2_t in self.conv2: + x2_t = conv2_t(x1) + x2 = x2 + self.gate(x2_t) + x3 = self.conv3(x2) + x3 = self.IN(x3) # IN inside residual + if self.downsample is not None: + identity = self.downsample(identity) + out = x3 + identity + return F.relu(out) + + +########## +# Network architecture +########## +class OSNet(nn.Module): + """Omni-Scale Network. + + Reference: + - Zhou et al. Omni-Scale Feature Learning for Person Re-Identification. ICCV, 2019. + - Zhou et al. Learning Generalisable Omni-Scale Representations + for Person Re-Identification. TPAMI, 2021. + """ + + def __init__( + self, + num_classes, + blocks, + layers, + channels, + feature_dim=512, + loss='softmax', + conv1_IN=False, + **kwargs + ): + super(OSNet, self).__init__() + num_blocks = len(blocks) + assert num_blocks == len(layers) + assert num_blocks == len(channels) - 1 + self.loss = loss + self.feature_dim = feature_dim + + # convolutional backbone + self.conv1 = ConvLayer( + 3, channels[0], 7, stride=2, padding=3, IN=conv1_IN + ) + self.maxpool = nn.MaxPool2d(3, stride=2, padding=1) + self.conv2 = self._make_layer( + blocks[0], layers[0], channels[0], channels[1] + ) + self.pool2 = nn.Sequential( + Conv1x1(channels[1], channels[1]), nn.AvgPool2d(2, stride=2) + ) + self.conv3 = self._make_layer( + blocks[1], layers[1], channels[1], channels[2] + ) + self.pool3 = nn.Sequential( + Conv1x1(channels[2], channels[2]), nn.AvgPool2d(2, stride=2) + ) + self.conv4 = self._make_layer( + blocks[2], layers[2], channels[2], channels[3] + ) + self.conv5 = Conv1x1(channels[3], channels[3]) + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + # fully connected layer + self.fc = self._construct_fc_layer( + self.feature_dim, channels[3], dropout_p=None + ) + # identity classification layer + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _make_layer(self, blocks, layer, in_channels, out_channels): + layers = [] + layers += [blocks[0](in_channels, out_channels)] + for i in range(1, len(blocks)): + layers += [blocks[i](out_channels, out_channels)] + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + if fc_dims is None or fc_dims < 0: + self.feature_dim = input_dim + return None + + if isinstance(fc_dims, int): + fc_dims = [fc_dims] + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU()) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.InstanceNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.maxpool(x) + x = self.conv2(x) + x = self.pool2(x) + x = self.conv3(x) + x = self.pool3(x) + x = self.conv4(x) + x = self.conv5(x) + return x + + def forward(self, x, return_featuremaps=False): + x = self.featuremaps(x) + if return_featuremaps: + return x + v = self.global_avgpool(x) + v = v.view(v.size(0), -1) + if self.fc is not None: + v = self.fc(v) + if not self.training: + return v + y = self.classifier(v) + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, key=''): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + import os + import errno + import gdown + from collections import OrderedDict + + def _get_torch_home(): + ENV_TORCH_HOME = 'TORCH_HOME' + ENV_XDG_CACHE_HOME = 'XDG_CACHE_HOME' + DEFAULT_CACHE_DIR = '~/.cache' + torch_home = os.path.expanduser( + os.getenv( + ENV_TORCH_HOME, + os.path.join( + os.getenv(ENV_XDG_CACHE_HOME, DEFAULT_CACHE_DIR), 'torch' + ) + ) + ) + return torch_home + + torch_home = _get_torch_home() + model_dir = os.path.join(torch_home, 'checkpoints') + try: + os.makedirs(model_dir) + except OSError as e: + if e.errno == errno.EEXIST: + # Directory already exists, ignore. + pass + else: + # Unexpected OSError, re-raise. + raise + filename = key + '_imagenet.pth' + cached_file = os.path.join(model_dir, filename) + + if not os.path.exists(cached_file): + gdown.download(pretrained_urls[key], cached_file, quiet=False) + + state_dict = torch.load(cached_file) + model_dict = model.state_dict() + new_state_dict = OrderedDict() + matched_layers, discarded_layers = [], [] + + for k, v in state_dict.items(): + if k.startswith('module.'): + k = k[7:] # discard module. + + if k in model_dict and model_dict[k].size() == v.size(): + new_state_dict[k] = v + matched_layers.append(k) + else: + discarded_layers.append(k) + + model_dict.update(new_state_dict) + model.load_state_dict(model_dict) + + if len(matched_layers) == 0: + warnings.warn( + 'The pretrained weights from "{}" cannot be loaded, ' + 'please check the key names manually ' + '(** ignored and continue **)'.format(cached_file) + ) + else: + print( + 'Successfully loaded imagenet pretrained weights from "{}"'. + format(cached_file) + ) + if len(discarded_layers) > 0: + print( + '** The following layers are discarded ' + 'due to unmatched keys or layer size: {}'. + format(discarded_layers) + ) + + +########## +# Instantiation +########## +def osnet_ain_x1_0( + num_classes=1000, pretrained=True, loss='softmax', **kwargs +): + model = OSNet( + num_classes, + blocks=[ + [OSBlockINin, OSBlockINin], [OSBlock, OSBlockINin], + [OSBlockINin, OSBlock] + ], + layers=[2, 2, 2], + channels=[64, 256, 384, 512], + loss=loss, + conv1_IN=True, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, key='osnet_ain_x1_0') + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py new file mode 100644 index 0000000000..8065574baa --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/pcb.py @@ -0,0 +1,361 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['pcb_p6', 'pcb_p4'] + +model_urls = { + 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', + 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', + 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', + 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', + 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', +} + + +def conv3x3(in_planes, out_planes, stride=1): + """3x3 convolution with padding""" + return nn.Conv2d( + in_planes, + out_planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = nn.BatchNorm2d(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(Bottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d( + planes, + planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d( + planes, planes * self.expansion, kernel_size=1, bias=False + ) + self.bn3 = nn.BatchNorm2d(planes * self.expansion) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class DimReduceLayer(nn.Module): + + def __init__(self, in_channels, out_channels, nonlinear): + super(DimReduceLayer, self).__init__() + layers = [] + layers.append( + nn.Conv2d( + in_channels, out_channels, 1, stride=1, padding=0, bias=False + ) + ) + layers.append(nn.BatchNorm2d(out_channels)) + + if nonlinear == 'relu': + layers.append(nn.ReLU(inplace=True)) + elif nonlinear == 'leakyrelu': + layers.append(nn.LeakyReLU(0.1)) + + self.layers = nn.Sequential(*layers) + + def forward(self, x): + return self.layers(x) + + +class PCB(nn.Module): + """Part-based Convolutional Baseline. + + Reference: + Sun et al. Beyond Part Models: Person Retrieval with Refined + Part Pooling (and A Strong Convolutional Baseline). ECCV 2018. + + Public keys: + - ``pcb_p4``: PCB with 4-part strips. + - ``pcb_p6``: PCB with 6-part strips. + """ + + def __init__( + self, + num_classes, + loss, + block, + layers, + parts=6, + reduced_dim=256, + nonlinear='relu', + **kwargs + ): + self.inplanes = 64 + super(PCB, self).__init__() + self.loss = loss + self.parts = parts + self.feature_dim = 512 * block.expansion + + # backbone network + self.conv1 = nn.Conv2d( + 3, 64, kernel_size=7, stride=2, padding=3, bias=False + ) + self.bn1 = nn.BatchNorm2d(64) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer(block, 128, layers[1], stride=2) + self.layer3 = self._make_layer(block, 256, layers[2], stride=2) + self.layer4 = self._make_layer(block, 512, layers[3], stride=1) + + # pcb layers + self.parts_avgpool = nn.AdaptiveAvgPool2d((self.parts, 1)) + self.dropout = nn.Dropout(p=0.5) + self.conv5 = DimReduceLayer( + 512 * block.expansion, reduced_dim, nonlinear=nonlinear + ) + self.feature_dim = reduced_dim + self.classifier = nn.ModuleList( + [ + nn.Linear(self.feature_dim, num_classes) + for _ in range(self.parts) + ] + ) + + self._init_params() + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d( + self.inplanes, + planes * block.expansion, + kernel_size=1, + stride=stride, + bias=False + ), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample)) + self.inplanes = planes * block.expansion + for i in range(1, blocks): + layers.append(block(self.inplanes, planes)) + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v_g = self.parts_avgpool(f) + + if not self.training: + v_g = F.normalize(v_g, p=2, dim=1) + return v_g.view(v_g.size(0), -1) + + v_g = self.dropout(v_g) + v_h = self.conv5(v_g) + + y = [] + for i in range(self.parts): + v_h_i = v_h[:, :, i, :] + v_h_i = v_h_i.view(v_h_i.size(0), -1) + y_i = self.classifier[i](v_h_i) + y.append(y_i) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + v_g = F.normalize(v_g, p=2, dim=1) + return y, v_g.view(v_g.size(0), -1) + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def pcb_p6(num_classes, loss='softmax', pretrained=True, **kwargs): + model = PCB( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=1, + parts=6, + reduced_dim=256, + nonlinear='relu', + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model + + +def pcb_p4(num_classes, loss='softmax', pretrained=True, **kwargs): + model = PCB( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=1, + parts=4, + reduced_dim=256, + nonlinear='relu', + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py new file mode 100644 index 0000000000..be87f3fbdf --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet.py @@ -0,0 +1,576 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. + +Code source: https://github.com/pytorch/vision +""" +from __future__ import division, absolute_import +import torch.utils.model_zoo as model_zoo +from torch import nn + +__all__ = [ + 'resnet18', 'resnet34', 'resnet50', 'resnet101', 'resnet152', + 'resnext50_32x4d', 'resnext101_32x8d', 'resnet50_fc512' +] + +model_urls = { + 'resnet18': + 'https://download.pytorch.org/models/resnet18-5c106cde.pth', + 'resnet34': + 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', + 'resnet50': + 'https://download.pytorch.org/models/resnet50-19c8e357.pth', + 'resnet101': + 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', + 'resnet152': + 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', + 'resnext50_32x4d': + 'https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth', + 'resnext101_32x8d': + 'https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth', +} + + +def conv3x3(in_planes, out_planes, stride=1, groups=1, dilation=1): + """3x3 convolution with padding""" + return nn.Conv2d( + in_planes, + out_planes, + kernel_size=3, + stride=stride, + padding=dilation, + groups=groups, + bias=False, + dilation=dilation + ) + + +def conv1x1(in_planes, out_planes, stride=1): + """1x1 convolution""" + return nn.Conv2d( + in_planes, out_planes, kernel_size=1, stride=stride, bias=False + ) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__( + self, + inplanes, + planes, + stride=1, + downsample=None, + groups=1, + base_width=64, + dilation=1, + norm_layer=None + ): + super(BasicBlock, self).__init__() + if norm_layer is None: + norm_layer = nn.BatchNorm2d + if groups != 1 or base_width != 64: + raise ValueError( + 'BasicBlock only supports groups=1 and base_width=64' + ) + if dilation > 1: + raise NotImplementedError( + "Dilation > 1 not supported in BasicBlock" + ) + # Both self.conv1 and self.downsample layers downsample the input when stride != 1 + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = norm_layer(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = norm_layer(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + identity = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + identity = self.downsample(x) + + out += identity + out = self.relu(out) + + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__( + self, + inplanes, + planes, + stride=1, + downsample=None, + groups=1, + base_width=64, + dilation=1, + norm_layer=None + ): + super(Bottleneck, self).__init__() + if norm_layer is None: + norm_layer = nn.BatchNorm2d + width = int(planes * (base_width/64.)) * groups + # Both self.conv2 and self.downsample layers downsample the input when stride != 1 + self.conv1 = conv1x1(inplanes, width) + self.bn1 = norm_layer(width) + self.conv2 = conv3x3(width, width, stride, groups, dilation) + self.bn2 = norm_layer(width) + self.conv3 = conv1x1(width, planes * self.expansion) + self.bn3 = norm_layer(planes * self.expansion) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + identity = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + identity = self.downsample(x) + + out += identity + out = self.relu(out) + + return out + + +class ResNet(nn.Module): + """Residual network. + + Reference: + - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. + - Xie et al. Aggregated Residual Transformations for Deep Neural Networks. CVPR 2017. + + Public keys: + - ``resnet18``: ResNet18. + - ``resnet34``: ResNet34. + - ``resnet50``: ResNet50. + - ``resnet101``: ResNet101. + - ``resnet152``: ResNet152. + - ``resnext50_32x4d``: ResNeXt50. + - ``resnext101_32x8d``: ResNeXt101. + - ``resnet50_fc512``: ResNet50 + FC. + """ + + def __init__( + self, + num_classes, + loss, + block, + layers, + zero_init_residual=False, + groups=1, + width_per_group=64, + replace_stride_with_dilation=None, + norm_layer=None, + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ): + super(ResNet, self).__init__() + if norm_layer is None: + norm_layer = nn.BatchNorm2d + self._norm_layer = norm_layer + self.loss = loss + self.feature_dim = 512 * block.expansion + self.inplanes = 64 + self.dilation = 1 + if replace_stride_with_dilation is None: + # each element in the tuple indicates if we should replace + # the 2x2 stride with a dilated convolution instead + replace_stride_with_dilation = [False, False, False] + if len(replace_stride_with_dilation) != 3: + raise ValueError( + "replace_stride_with_dilation should be None " + "or a 3-element tuple, got {}". + format(replace_stride_with_dilation) + ) + self.groups = groups + self.base_width = width_per_group + self.conv1 = nn.Conv2d( + 3, self.inplanes, kernel_size=7, stride=2, padding=3, bias=False + ) + self.bn1 = norm_layer(self.inplanes) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer( + block, + 128, + layers[1], + stride=2, + dilate=replace_stride_with_dilation[0] + ) + self.layer3 = self._make_layer( + block, + 256, + layers[2], + stride=2, + dilate=replace_stride_with_dilation[1] + ) + self.layer4 = self._make_layer( + block, + 512, + layers[3], + stride=last_stride, + dilate=replace_stride_with_dilation[2] + ) + self.global_avgpool = nn.AdaptiveAvgPool2d((1, 1)) + self.fc = self._construct_fc_layer( + fc_dims, 512 * block.expansion, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + # Zero-initialize the last BN in each residual branch, + # so that the residual branch starts with zeros, and each residual block behaves like an identity. + # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677 + if zero_init_residual: + for m in self.modules(): + if isinstance(m, Bottleneck): + nn.init.constant_(m.bn3.weight, 0) + elif isinstance(m, BasicBlock): + nn.init.constant_(m.bn2.weight, 0) + + def _make_layer(self, block, planes, blocks, stride=1, dilate=False): + norm_layer = self._norm_layer + downsample = None + previous_dilation = self.dilation + if dilate: + self.dilation *= stride + stride = 1 + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + conv1x1(self.inplanes, planes * block.expansion, stride), + norm_layer(planes * block.expansion), + ) + + layers = [] + layers.append( + block( + self.inplanes, planes, stride, downsample, self.groups, + self.base_width, previous_dilation, norm_layer + ) + ) + self.inplanes = planes * block.expansion + for _ in range(1, blocks): + layers.append( + block( + self.inplanes, + planes, + groups=self.groups, + base_width=self.base_width, + dilation=self.dilation, + norm_layer=norm_layer + ) + ) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +"""ResNet""" + + +def resnet18(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=BasicBlock, + layers=[2, 2, 2, 2], + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet18']) + return model + + +def resnet34(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=BasicBlock, + layers=[3, 4, 6, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet34']) + return model + + +def resnet50(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model + + +def resnet101(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 23, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet101']) + return model + + +def resnet152(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 8, 36, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet152']) + return model + + +"""ResNeXt""" + + +def resnext50_32x4d(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + groups=32, + width_per_group=4, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnext50_32x4d']) + return model + + +def resnext101_32x8d(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 23, 3], + last_stride=2, + fc_dims=None, + dropout_p=None, + groups=32, + width_per_group=8, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnext101_32x8d']) + return model + + +""" +ResNet + FC +""" + + +def resnet50_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNet( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=1, + fc_dims=[512], + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py new file mode 100644 index 0000000000..579f5ed22b --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_a.py @@ -0,0 +1,334 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Credit to https://github.com/XingangPan/IBN-Net. +""" +from __future__ import division, absolute_import +import math +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['resnet50_ibn_a'] + +model_urls = { + 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', + 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', + 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', +} + + +def conv3x3(in_planes, out_planes, stride=1): + "3x3 convolution with padding" + return nn.Conv2d( + in_planes, + out_planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = nn.BatchNorm2d(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class IBN(nn.Module): + + def __init__(self, planes): + super(IBN, self).__init__() + half1 = int(planes / 2) + self.half = half1 + half2 = planes - half1 + self.IN = nn.InstanceNorm2d(half1, affine=True) + self.BN = nn.BatchNorm2d(half2) + + def forward(self, x): + split = torch.split(x, self.half, 1) + out1 = self.IN(split[0].contiguous()) + out2 = self.BN(split[1].contiguous()) + out = torch.cat((out1, out2), 1) + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__(self, inplanes, planes, ibn=False, stride=1, downsample=None): + super(Bottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) + if ibn: + self.bn1 = IBN(planes) + else: + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d( + planes, + planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d( + planes, planes * self.expansion, kernel_size=1, bias=False + ) + self.bn3 = nn.BatchNorm2d(planes * self.expansion) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class ResNet(nn.Module): + """Residual network + IBN layer. + + Reference: + - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. + - Pan et al. Two at Once: Enhancing Learning and Generalization + Capacities via IBN-Net. ECCV 2018. + """ + + def __init__( + self, + block, + layers, + num_classes=1000, + loss='softmax', + fc_dims=None, + dropout_p=None, + **kwargs + ): + scale = 64 + self.inplanes = scale + super(ResNet, self).__init__() + self.loss = loss + self.feature_dim = scale * 8 * block.expansion + + self.conv1 = nn.Conv2d( + 3, scale, kernel_size=7, stride=2, padding=3, bias=False + ) + self.bn1 = nn.BatchNorm2d(scale) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = self._make_layer(block, scale, layers[0]) + self.layer2 = self._make_layer(block, scale * 2, layers[1], stride=2) + self.layer3 = self._make_layer(block, scale * 4, layers[2], stride=2) + self.layer4 = self._make_layer(block, scale * 8, layers[3], stride=2) + self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) + self.fc = self._construct_fc_layer( + fc_dims, scale * 8 * block.expansion, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + for m in self.modules(): + if isinstance(m, nn.Conv2d): + n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels + m.weight.data.normal_(0, math.sqrt(2. / n)) + elif isinstance(m, nn.BatchNorm2d): + m.weight.data.fill_(1) + m.bias.data.zero_() + elif isinstance(m, nn.InstanceNorm2d): + m.weight.data.fill_(1) + m.bias.data.zero_() + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d( + self.inplanes, + planes * block.expansion, + kernel_size=1, + stride=stride, + bias=False + ), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + ibn = True + if planes == 512: + ibn = False + layers.append(block(self.inplanes, planes, ibn, stride, downsample)) + self.inplanes = planes * block.expansion + for i in range(1, blocks): + layers.append(block(self.inplanes, planes, ibn)) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.avgpool(f) + v = v.view(v.size(0), -1) + if self.fc is not None: + v = self.fc(v) + if not self.training: + return v + y = self.classifier(v) + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def resnet50_ibn_a(num_classes, loss='softmax', pretrained=False, **kwargs): + model = ResNet( + Bottleneck, [3, 4, 6, 3], num_classes=num_classes, loss=loss, **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py new file mode 100644 index 0000000000..7026fb9e3d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnet_ibn_b.py @@ -0,0 +1,319 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Credit to https://github.com/XingangPan/IBN-Net. +""" +from __future__ import division, absolute_import +import math +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['resnet50_ibn_b'] + +model_urls = { + 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', + 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', + 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', +} + + +def conv3x3(in_planes, out_planes, stride=1): + "3x3 convolution with padding" + return nn.Conv2d( + in_planes, + out_planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = nn.BatchNorm2d(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__(self, inplanes, planes, stride=1, downsample=None, IN=False): + super(Bottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d( + planes, + planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d( + planes, planes * self.expansion, kernel_size=1, bias=False + ) + self.bn3 = nn.BatchNorm2d(planes * self.expansion) + self.IN = None + if IN: + self.IN = nn.InstanceNorm2d(planes * 4, affine=True) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + if self.IN is not None: + out = self.IN(out) + out = self.relu(out) + + return out + + +class ResNet(nn.Module): + """Residual network + IBN layer. + + Reference: + - He et al. Deep Residual Learning for Image Recognition. CVPR 2016. + - Pan et al. Two at Once: Enhancing Learning and Generalization + Capacities via IBN-Net. ECCV 2018. + """ + + def __init__( + self, + block, + layers, + num_classes=1000, + loss='softmax', + fc_dims=None, + dropout_p=None, + **kwargs + ): + scale = 64 + self.inplanes = scale + super(ResNet, self).__init__() + self.loss = loss + self.feature_dim = scale * 8 * block.expansion + + self.conv1 = nn.Conv2d( + 3, scale, kernel_size=7, stride=2, padding=3, bias=False + ) + self.bn1 = nn.InstanceNorm2d(scale, affine=True) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = self._make_layer( + block, scale, layers[0], stride=1, IN=True + ) + self.layer2 = self._make_layer( + block, scale * 2, layers[1], stride=2, IN=True + ) + self.layer3 = self._make_layer(block, scale * 4, layers[2], stride=2) + self.layer4 = self._make_layer(block, scale * 8, layers[3], stride=2) + self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) + self.fc = self._construct_fc_layer( + fc_dims, scale * 8 * block.expansion, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + for m in self.modules(): + if isinstance(m, nn.Conv2d): + n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels + m.weight.data.normal_(0, math.sqrt(2. / n)) + elif isinstance(m, nn.BatchNorm2d): + m.weight.data.fill_(1) + m.bias.data.zero_() + elif isinstance(m, nn.InstanceNorm2d): + m.weight.data.fill_(1) + m.bias.data.zero_() + + def _make_layer(self, block, planes, blocks, stride=1, IN=False): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d( + self.inplanes, + planes * block.expansion, + kernel_size=1, + stride=stride, + bias=False + ), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample)) + self.inplanes = planes * block.expansion + for i in range(1, blocks - 1): + layers.append(block(self.inplanes, planes)) + layers.append(block(self.inplanes, planes, IN=IN)) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.avgpool(f) + v = v.view(v.size(0), -1) + if self.fc is not None: + v = self.fc(v) + if not self.training: + return v + y = self.classifier(v) + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def resnet50_ibn_b(num_classes, loss='softmax', pretrained=False, **kwargs): + model = ResNet( + Bottleneck, [3, 4, 6, 3], num_classes=num_classes, loss=loss, **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py new file mode 100644 index 0000000000..87130a2fb2 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/resnetmid.py @@ -0,0 +1,354 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.utils.model_zoo as model_zoo +from torch import nn + +__all__ = ['resnet50mid'] + +model_urls = { + 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', + 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', + 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', + 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', + 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', +} + + +def conv3x3(in_planes, out_planes, stride=1): + """3x3 convolution with padding""" + return nn.Conv2d( + in_planes, + out_planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + + +class BasicBlock(nn.Module): + expansion = 1 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(BasicBlock, self).__init__() + self.conv1 = conv3x3(inplanes, planes, stride) + self.bn1 = nn.BatchNorm2d(planes) + self.relu = nn.ReLU(inplace=True) + self.conv2 = conv3x3(planes, planes) + self.bn2 = nn.BatchNorm2d(planes) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class Bottleneck(nn.Module): + expansion = 4 + + def __init__(self, inplanes, planes, stride=1, downsample=None): + super(Bottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d( + planes, + planes, + kernel_size=3, + stride=stride, + padding=1, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d( + planes, planes * self.expansion, kernel_size=1, bias=False + ) + self.bn3 = nn.BatchNorm2d(planes * self.expansion) + self.relu = nn.ReLU(inplace=True) + self.downsample = downsample + self.stride = stride + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out += residual + out = self.relu(out) + + return out + + +class ResNetMid(nn.Module): + """Residual network + mid-level features. + + Reference: + Yu et al. The Devil is in the Middle: Exploiting Mid-level Representations for + Cross-Domain Instance Matching. arXiv:1711.08106. + + Public keys: + - ``resnet50mid``: ResNet50 + mid-level feature fusion. + """ + + def __init__( + self, + num_classes, + loss, + block, + layers, + last_stride=2, + fc_dims=None, + **kwargs + ): + self.inplanes = 64 + super(ResNetMid, self).__init__() + self.loss = loss + self.feature_dim = 512 * block.expansion + + # backbone network + self.conv1 = nn.Conv2d( + 3, 64, kernel_size=7, stride=2, padding=3, bias=False + ) + self.bn1 = nn.BatchNorm2d(64) + self.relu = nn.ReLU(inplace=True) + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + self.layer1 = self._make_layer(block, 64, layers[0]) + self.layer2 = self._make_layer(block, 128, layers[1], stride=2) + self.layer3 = self._make_layer(block, 256, layers[2], stride=2) + self.layer4 = self._make_layer( + block, 512, layers[3], stride=last_stride + ) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + assert fc_dims is not None + self.fc_fusion = self._construct_fc_layer( + fc_dims, 512 * block.expansion * 2 + ) + self.feature_dim += 512 * block.expansion + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _make_layer(self, block, planes, blocks, stride=1): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d( + self.inplanes, + planes * block.expansion, + kernel_size=1, + stride=stride, + bias=False + ), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append(block(self.inplanes, planes, stride, downsample)) + self.inplanes = planes * block.expansion + for i in range(1, blocks): + layers.append(block(self.inplanes, planes)) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.bn1(x) + x = self.relu(x) + x = self.maxpool(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x4a = self.layer4[0](x) + x4b = self.layer4[1](x4a) + x4c = self.layer4[2](x4b) + return x4a, x4b, x4c + + def forward(self, x): + x4a, x4b, x4c = self.featuremaps(x) + + v4a = self.global_avgpool(x4a) + v4b = self.global_avgpool(x4b) + v4c = self.global_avgpool(x4c) + v4ab = torch.cat([v4a, v4b], 1) + v4ab = v4ab.view(v4ab.size(0), -1) + v4ab = self.fc_fusion(v4ab) + v4c = v4c.view(v4c.size(0), -1) + v = torch.cat([v4ab, v4c], 1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +""" +Residual network configurations: +-- +resnet18: block=BasicBlock, layers=[2, 2, 2, 2] +resnet34: block=BasicBlock, layers=[3, 4, 6, 3] +resnet50: block=Bottleneck, layers=[3, 4, 6, 3] +resnet101: block=Bottleneck, layers=[3, 4, 23, 3] +resnet152: block=Bottleneck, layers=[3, 8, 36, 3] +""" + + +def resnet50mid(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ResNetMid( + num_classes=num_classes, + loss=loss, + block=Bottleneck, + layers=[3, 4, 6, 3], + last_stride=2, + fc_dims=[1024], + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['resnet50']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py new file mode 100644 index 0000000000..ef38f0e56f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/senet.py @@ -0,0 +1,735 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import math +from collections import OrderedDict +import torch.nn as nn +from torch.utils import model_zoo + +__all__ = [ + 'senet154', 'se_resnet50', 'se_resnet101', 'se_resnet152', + 'se_resnext50_32x4d', 'se_resnext101_32x4d', 'se_resnet50_fc512' +] +""" +Code imported from https://github.com/Cadene/pretrained-models.pytorch +""" + +pretrained_settings = { + 'senet154': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/senet154-c7b49a05.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, + 'se_resnet50': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet50-ce0d4300.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, + 'se_resnet101': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet101-7e38fcc6.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, + 'se_resnet152': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnet152-d17c99b7.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, + 'se_resnext50_32x4d': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnext50_32x4d-a260b3a4.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, + 'se_resnext101_32x4d': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/se_resnext101_32x4d-3b2fe3d8.pth', + 'input_space': 'RGB', + 'input_size': [3, 224, 224], + 'input_range': [0, 1], + 'mean': [0.485, 0.456, 0.406], + 'std': [0.229, 0.224, 0.225], + 'num_classes': 1000 + } + }, +} + + +class SEModule(nn.Module): + + def __init__(self, channels, reduction): + super(SEModule, self).__init__() + self.avg_pool = nn.AdaptiveAvgPool2d(1) + self.fc1 = nn.Conv2d( + channels, channels // reduction, kernel_size=1, padding=0 + ) + self.relu = nn.ReLU(inplace=True) + self.fc2 = nn.Conv2d( + channels // reduction, channels, kernel_size=1, padding=0 + ) + self.sigmoid = nn.Sigmoid() + + def forward(self, x): + module_input = x + x = self.avg_pool(x) + x = self.fc1(x) + x = self.relu(x) + x = self.fc2(x) + x = self.sigmoid(x) + return module_input * x + + +class Bottleneck(nn.Module): + """ + Base class for bottlenecks that implements `forward()` method. + """ + + def forward(self, x): + residual = x + + out = self.conv1(x) + out = self.bn1(out) + out = self.relu(out) + + out = self.conv2(out) + out = self.bn2(out) + out = self.relu(out) + + out = self.conv3(out) + out = self.bn3(out) + + if self.downsample is not None: + residual = self.downsample(x) + + out = self.se_module(out) + residual + out = self.relu(out) + + return out + + +class SEBottleneck(Bottleneck): + """ + Bottleneck for SENet154. + """ + expansion = 4 + + def __init__( + self, inplanes, planes, groups, reduction, stride=1, downsample=None + ): + super(SEBottleneck, self).__init__() + self.conv1 = nn.Conv2d(inplanes, planes * 2, kernel_size=1, bias=False) + self.bn1 = nn.BatchNorm2d(planes * 2) + self.conv2 = nn.Conv2d( + planes * 2, + planes * 4, + kernel_size=3, + stride=stride, + padding=1, + groups=groups, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes * 4) + self.conv3 = nn.Conv2d( + planes * 4, planes * 4, kernel_size=1, bias=False + ) + self.bn3 = nn.BatchNorm2d(planes * 4) + self.relu = nn.ReLU(inplace=True) + self.se_module = SEModule(planes * 4, reduction=reduction) + self.downsample = downsample + self.stride = stride + + +class SEResNetBottleneck(Bottleneck): + """ + ResNet bottleneck with a Squeeze-and-Excitation module. It follows Caffe + implementation and uses `stride=stride` in `conv1` and not in `conv2` + (the latter is used in the torchvision implementation of ResNet). + """ + expansion = 4 + + def __init__( + self, inplanes, planes, groups, reduction, stride=1, downsample=None + ): + super(SEResNetBottleneck, self).__init__() + self.conv1 = nn.Conv2d( + inplanes, planes, kernel_size=1, bias=False, stride=stride + ) + self.bn1 = nn.BatchNorm2d(planes) + self.conv2 = nn.Conv2d( + planes, + planes, + kernel_size=3, + padding=1, + groups=groups, + bias=False + ) + self.bn2 = nn.BatchNorm2d(planes) + self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False) + self.bn3 = nn.BatchNorm2d(planes * 4) + self.relu = nn.ReLU(inplace=True) + self.se_module = SEModule(planes * 4, reduction=reduction) + self.downsample = downsample + self.stride = stride + + +class SEResNeXtBottleneck(Bottleneck): + """ResNeXt bottleneck type C with a Squeeze-and-Excitation module""" + expansion = 4 + + def __init__( + self, + inplanes, + planes, + groups, + reduction, + stride=1, + downsample=None, + base_width=4 + ): + super(SEResNeXtBottleneck, self).__init__() + width = int(math.floor(planes * (base_width/64.)) * groups) + self.conv1 = nn.Conv2d( + inplanes, width, kernel_size=1, bias=False, stride=1 + ) + self.bn1 = nn.BatchNorm2d(width) + self.conv2 = nn.Conv2d( + width, + width, + kernel_size=3, + stride=stride, + padding=1, + groups=groups, + bias=False + ) + self.bn2 = nn.BatchNorm2d(width) + self.conv3 = nn.Conv2d(width, planes * 4, kernel_size=1, bias=False) + self.bn3 = nn.BatchNorm2d(planes * 4) + self.relu = nn.ReLU(inplace=True) + self.se_module = SEModule(planes * 4, reduction=reduction) + self.downsample = downsample + self.stride = stride + + +class SENet(nn.Module): + """Squeeze-and-excitation network. + + Reference: + Hu et al. Squeeze-and-Excitation Networks. CVPR 2018. + + Public keys: + - ``senet154``: SENet154. + - ``se_resnet50``: ResNet50 + SE. + - ``se_resnet101``: ResNet101 + SE. + - ``se_resnet152``: ResNet152 + SE. + - ``se_resnext50_32x4d``: ResNeXt50 (groups=32, width=4) + SE. + - ``se_resnext101_32x4d``: ResNeXt101 (groups=32, width=4) + SE. + - ``se_resnet50_fc512``: (ResNet50 + SE) + FC. + """ + + def __init__( + self, + num_classes, + loss, + block, + layers, + groups, + reduction, + dropout_p=0.2, + inplanes=128, + input_3x3=True, + downsample_kernel_size=3, + downsample_padding=1, + last_stride=2, + fc_dims=None, + **kwargs + ): + """ + Parameters + ---------- + block (nn.Module): Bottleneck class. + - For SENet154: SEBottleneck + - For SE-ResNet models: SEResNetBottleneck + - For SE-ResNeXt models: SEResNeXtBottleneck + layers (list of ints): Number of residual blocks for 4 layers of the + network (layer1...layer4). + groups (int): Number of groups for the 3x3 convolution in each + bottleneck block. + - For SENet154: 64 + - For SE-ResNet models: 1 + - For SE-ResNeXt models: 32 + reduction (int): Reduction ratio for Squeeze-and-Excitation modules. + - For all models: 16 + dropout_p (float or None): Drop probability for the Dropout layer. + If `None` the Dropout layer is not used. + - For SENet154: 0.2 + - For SE-ResNet models: None + - For SE-ResNeXt models: None + inplanes (int): Number of input channels for layer1. + - For SENet154: 128 + - For SE-ResNet models: 64 + - For SE-ResNeXt models: 64 + input_3x3 (bool): If `True`, use three 3x3 convolutions instead of + a single 7x7 convolution in layer0. + - For SENet154: True + - For SE-ResNet models: False + - For SE-ResNeXt models: False + downsample_kernel_size (int): Kernel size for downsampling convolutions + in layer2, layer3 and layer4. + - For SENet154: 3 + - For SE-ResNet models: 1 + - For SE-ResNeXt models: 1 + downsample_padding (int): Padding for downsampling convolutions in + layer2, layer3 and layer4. + - For SENet154: 1 + - For SE-ResNet models: 0 + - For SE-ResNeXt models: 0 + num_classes (int): Number of outputs in `classifier` layer. + """ + super(SENet, self).__init__() + self.inplanes = inplanes + self.loss = loss + + if input_3x3: + layer0_modules = [ + ( + 'conv1', + nn.Conv2d(3, 64, 3, stride=2, padding=1, bias=False) + ), + ('bn1', nn.BatchNorm2d(64)), + ('relu1', nn.ReLU(inplace=True)), + ( + 'conv2', + nn.Conv2d(64, 64, 3, stride=1, padding=1, bias=False) + ), + ('bn2', nn.BatchNorm2d(64)), + ('relu2', nn.ReLU(inplace=True)), + ( + 'conv3', + nn.Conv2d( + 64, inplanes, 3, stride=1, padding=1, bias=False + ) + ), + ('bn3', nn.BatchNorm2d(inplanes)), + ('relu3', nn.ReLU(inplace=True)), + ] + else: + layer0_modules = [ + ( + 'conv1', + nn.Conv2d( + 3, + inplanes, + kernel_size=7, + stride=2, + padding=3, + bias=False + ) + ), + ('bn1', nn.BatchNorm2d(inplanes)), + ('relu1', nn.ReLU(inplace=True)), + ] + # To preserve compatibility with Caffe weights `ceil_mode=True` + # is used instead of `padding=1`. + layer0_modules.append( + ('pool', nn.MaxPool2d(3, stride=2, ceil_mode=True)) + ) + self.layer0 = nn.Sequential(OrderedDict(layer0_modules)) + self.layer1 = self._make_layer( + block, + planes=64, + blocks=layers[0], + groups=groups, + reduction=reduction, + downsample_kernel_size=1, + downsample_padding=0 + ) + self.layer2 = self._make_layer( + block, + planes=128, + blocks=layers[1], + stride=2, + groups=groups, + reduction=reduction, + downsample_kernel_size=downsample_kernel_size, + downsample_padding=downsample_padding + ) + self.layer3 = self._make_layer( + block, + planes=256, + blocks=layers[2], + stride=2, + groups=groups, + reduction=reduction, + downsample_kernel_size=downsample_kernel_size, + downsample_padding=downsample_padding + ) + self.layer4 = self._make_layer( + block, + planes=512, + blocks=layers[3], + stride=last_stride, + groups=groups, + reduction=reduction, + downsample_kernel_size=downsample_kernel_size, + downsample_padding=downsample_padding + ) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc = self._construct_fc_layer( + fc_dims, 512 * block.expansion, dropout_p + ) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + def _make_layer( + self, + block, + planes, + blocks, + groups, + reduction, + stride=1, + downsample_kernel_size=1, + downsample_padding=0 + ): + downsample = None + if stride != 1 or self.inplanes != planes * block.expansion: + downsample = nn.Sequential( + nn.Conv2d( + self.inplanes, + planes * block.expansion, + kernel_size=downsample_kernel_size, + stride=stride, + padding=downsample_padding, + bias=False + ), + nn.BatchNorm2d(planes * block.expansion), + ) + + layers = [] + layers.append( + block( + self.inplanes, planes, groups, reduction, stride, downsample + ) + ) + self.inplanes = planes * block.expansion + for i in range(1, blocks): + layers.append(block(self.inplanes, planes, groups, reduction)) + + return nn.Sequential(*layers) + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """ + Construct fully connected layer + + - fc_dims (list or tuple): dimensions of fc layers, if None, + no fc layers are constructed + - input_dim (int): input dimension + - dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def featuremaps(self, x): + x = self.layer0(x) + x = self.layer1(x) + x = self.layer2(x) + x = self.layer3(x) + x = self.layer4(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def senet154(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEBottleneck, + layers=[3, 8, 36, 3], + groups=64, + reduction=16, + dropout_p=0.2, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['senet154']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model + + +def se_resnet50(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNetBottleneck, + layers=[3, 4, 6, 3], + groups=1, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnet50']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model + + +def se_resnet50_fc512(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNetBottleneck, + layers=[3, 4, 6, 3], + groups=1, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=1, + fc_dims=[512], + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnet50']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model + + +def se_resnet101(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNetBottleneck, + layers=[3, 4, 23, 3], + groups=1, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnet101']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model + + +def se_resnet152(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNetBottleneck, + layers=[3, 8, 36, 3], + groups=1, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnet152']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model + + +def se_resnext50_32x4d(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNeXtBottleneck, + layers=[3, 4, 6, 3], + groups=32, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnext50_32x4d']['imagenet']['url' + ] + init_pretrained_weights(model, model_url) + return model + + +def se_resnext101_32x4d( + num_classes, loss='softmax', pretrained=True, **kwargs +): + model = SENet( + num_classes=num_classes, + loss=loss, + block=SEResNeXtBottleneck, + layers=[3, 4, 23, 3], + groups=32, + reduction=16, + dropout_p=None, + inplanes=64, + input_3x3=False, + downsample_kernel_size=1, + downsample_padding=0, + last_stride=2, + fc_dims=None, + **kwargs + ) + if pretrained: + model_url = pretrained_settings['se_resnext101_32x4d']['imagenet'][ + 'url'] + init_pretrained_weights(model, model_url) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py new file mode 100644 index 0000000000..3e05ff7bc5 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenet.py @@ -0,0 +1,245 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch +import torch.utils.model_zoo as model_zoo +from torch import nn +from torch.nn import functional as F + +__all__ = ['shufflenet'] + +model_urls = { + # training epoch = 90, top1 = 61.8 + 'imagenet': + 'https://mega.nz/#!RDpUlQCY!tr_5xBEkelzDjveIYBBcGcovNCOrgfiJO9kiidz9fZM', +} + + +class ChannelShuffle(nn.Module): + + def __init__(self, num_groups): + super(ChannelShuffle, self).__init__() + self.g = num_groups + + def forward(self, x): + b, c, h, w = x.size() + n = c // self.g + # reshape + x = x.view(b, self.g, n, h, w) + # transpose + x = x.permute(0, 2, 1, 3, 4).contiguous() + # flatten + x = x.view(b, c, h, w) + return x + + +class Bottleneck(nn.Module): + + def __init__( + self, + in_channels, + out_channels, + stride, + num_groups, + group_conv1x1=True + ): + super(Bottleneck, self).__init__() + assert stride in [1, 2], 'Warning: stride must be either 1 or 2' + self.stride = stride + mid_channels = out_channels // 4 + if stride == 2: + out_channels -= in_channels + # group conv is not applied to first conv1x1 at stage 2 + num_groups_conv1x1 = num_groups if group_conv1x1 else 1 + self.conv1 = nn.Conv2d( + in_channels, + mid_channels, + 1, + groups=num_groups_conv1x1, + bias=False + ) + self.bn1 = nn.BatchNorm2d(mid_channels) + self.shuffle1 = ChannelShuffle(num_groups) + self.conv2 = nn.Conv2d( + mid_channels, + mid_channels, + 3, + stride=stride, + padding=1, + groups=mid_channels, + bias=False + ) + self.bn2 = nn.BatchNorm2d(mid_channels) + self.conv3 = nn.Conv2d( + mid_channels, out_channels, 1, groups=num_groups, bias=False + ) + self.bn3 = nn.BatchNorm2d(out_channels) + if stride == 2: + self.shortcut = nn.AvgPool2d(3, stride=2, padding=1) + + def forward(self, x): + out = F.relu(self.bn1(self.conv1(x))) + out = self.shuffle1(out) + out = self.bn2(self.conv2(out)) + out = self.bn3(self.conv3(out)) + if self.stride == 2: + res = self.shortcut(x) + out = F.relu(torch.cat([res, out], 1)) + else: + out = F.relu(x + out) + return out + + +# configuration of (num_groups: #out_channels) based on Table 1 in the paper +cfg = { + 1: [144, 288, 576], + 2: [200, 400, 800], + 3: [240, 480, 960], + 4: [272, 544, 1088], + 8: [384, 768, 1536], +} + + +class ShuffleNet(nn.Module): + """ShuffleNet. + + Reference: + Zhang et al. ShuffleNet: An Extremely Efficient Convolutional Neural + Network for Mobile Devices. CVPR 2018. + + Public keys: + - ``shufflenet``: ShuffleNet (groups=3). + """ + + def __init__(self, num_classes, loss='softmax', num_groups=3, **kwargs): + super(ShuffleNet, self).__init__() + self.loss = loss + + self.conv1 = nn.Sequential( + nn.Conv2d(3, 24, 3, stride=2, padding=1, bias=False), + nn.BatchNorm2d(24), + nn.ReLU(), + nn.MaxPool2d(3, stride=2, padding=1), + ) + + self.stage2 = nn.Sequential( + Bottleneck( + 24, cfg[num_groups][0], 2, num_groups, group_conv1x1=False + ), + Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), + Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), + Bottleneck(cfg[num_groups][0], cfg[num_groups][0], 1, num_groups), + ) + + self.stage3 = nn.Sequential( + Bottleneck(cfg[num_groups][0], cfg[num_groups][1], 2, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + Bottleneck(cfg[num_groups][1], cfg[num_groups][1], 1, num_groups), + ) + + self.stage4 = nn.Sequential( + Bottleneck(cfg[num_groups][1], cfg[num_groups][2], 2, num_groups), + Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), + Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), + Bottleneck(cfg[num_groups][2], cfg[num_groups][2], 1, num_groups), + ) + + self.classifier = nn.Linear(cfg[num_groups][2], num_classes) + self.feat_dim = cfg[num_groups][2] + + def forward(self, x): + x = self.conv1(x) + x = self.stage2(x) + x = self.stage3(x) + x = self.stage4(x) + x = F.avg_pool2d(x, x.size()[2:]).view(x.size(0), -1) + + if not self.training: + return x + + y = self.classifier(x) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, x + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def shufflenet(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ShuffleNet(num_classes, loss, **kwargs) + if pretrained: + # init_pretrained_weights(model, model_urls['imagenet']) + import warnings + warnings.warn( + 'The imagenet pretrained weights need to be manually downloaded from {}' + .format(model_urls['imagenet']) + ) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py new file mode 100644 index 0000000000..2b9fa4d403 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/shufflenetv2.py @@ -0,0 +1,307 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code source: https://github.com/pytorch/vision +""" +from __future__ import division, absolute_import +import torch +import torch.utils.model_zoo as model_zoo +from torch import nn + +__all__ = [ + 'shufflenet_v2_x0_5', 'shufflenet_v2_x1_0', 'shufflenet_v2_x1_5', + 'shufflenet_v2_x2_0' +] + +model_urls = { + 'shufflenetv2_x0.5': + 'https://download.pytorch.org/models/shufflenetv2_x0.5-f707e7126e.pth', + 'shufflenetv2_x1.0': + 'https://download.pytorch.org/models/shufflenetv2_x1-5666bf0f80.pth', + 'shufflenetv2_x1.5': None, + 'shufflenetv2_x2.0': None, +} + + +def channel_shuffle(x, groups): + batchsize, num_channels, height, width = x.data.size() + channels_per_group = num_channels // groups + + # reshape + x = x.view(batchsize, groups, channels_per_group, height, width) + + x = torch.transpose(x, 1, 2).contiguous() + + # flatten + x = x.view(batchsize, -1, height, width) + + return x + + +class InvertedResidual(nn.Module): + + def __init__(self, inp, oup, stride): + super(InvertedResidual, self).__init__() + + if not (1 <= stride <= 3): + raise ValueError('illegal stride value') + self.stride = stride + + branch_features = oup // 2 + assert (self.stride != 1) or (inp == branch_features << 1) + + if self.stride > 1: + self.branch1 = nn.Sequential( + self.depthwise_conv( + inp, inp, kernel_size=3, stride=self.stride, padding=1 + ), + nn.BatchNorm2d(inp), + nn.Conv2d( + inp, + branch_features, + kernel_size=1, + stride=1, + padding=0, + bias=False + ), + nn.BatchNorm2d(branch_features), + nn.ReLU(inplace=True), + ) + + self.branch2 = nn.Sequential( + nn.Conv2d( + inp if (self.stride > 1) else branch_features, + branch_features, + kernel_size=1, + stride=1, + padding=0, + bias=False + ), + nn.BatchNorm2d(branch_features), + nn.ReLU(inplace=True), + self.depthwise_conv( + branch_features, + branch_features, + kernel_size=3, + stride=self.stride, + padding=1 + ), + nn.BatchNorm2d(branch_features), + nn.Conv2d( + branch_features, + branch_features, + kernel_size=1, + stride=1, + padding=0, + bias=False + ), + nn.BatchNorm2d(branch_features), + nn.ReLU(inplace=True), + ) + + @staticmethod + def depthwise_conv(i, o, kernel_size, stride=1, padding=0, bias=False): + return nn.Conv2d( + i, o, kernel_size, stride, padding, bias=bias, groups=i + ) + + def forward(self, x): + if self.stride == 1: + x1, x2 = x.chunk(2, dim=1) + out = torch.cat((x1, self.branch2(x2)), dim=1) + else: + out = torch.cat((self.branch1(x), self.branch2(x)), dim=1) + + out = channel_shuffle(out, 2) + + return out + + +class ShuffleNetV2(nn.Module): + """ShuffleNetV2. + + Reference: + Ma et al. ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. ECCV 2018. + + Public keys: + - ``shufflenet_v2_x0_5``: ShuffleNetV2 x0.5. + - ``shufflenet_v2_x1_0``: ShuffleNetV2 x1.0. + - ``shufflenet_v2_x1_5``: ShuffleNetV2 x1.5. + - ``shufflenet_v2_x2_0``: ShuffleNetV2 x2.0. + """ + + def __init__( + self, num_classes, loss, stages_repeats, stages_out_channels, **kwargs + ): + super(ShuffleNetV2, self).__init__() + self.loss = loss + + if len(stages_repeats) != 3: + raise ValueError( + 'expected stages_repeats as list of 3 positive ints' + ) + if len(stages_out_channels) != 5: + raise ValueError( + 'expected stages_out_channels as list of 5 positive ints' + ) + self._stage_out_channels = stages_out_channels + + input_channels = 3 + output_channels = self._stage_out_channels[0] + self.conv1 = nn.Sequential( + nn.Conv2d(input_channels, output_channels, 3, 2, 1, bias=False), + nn.BatchNorm2d(output_channels), + nn.ReLU(inplace=True), + ) + input_channels = output_channels + + self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) + + stage_names = ['stage{}'.format(i) for i in [2, 3, 4]] + for name, repeats, output_channels in zip( + stage_names, stages_repeats, self._stage_out_channels[1:] + ): + seq = [InvertedResidual(input_channels, output_channels, 2)] + for i in range(repeats - 1): + seq.append( + InvertedResidual(output_channels, output_channels, 1) + ) + setattr(self, name, nn.Sequential(*seq)) + input_channels = output_channels + + output_channels = self._stage_out_channels[-1] + self.conv5 = nn.Sequential( + nn.Conv2d(input_channels, output_channels, 1, 1, 0, bias=False), + nn.BatchNorm2d(output_channels), + nn.ReLU(inplace=True), + ) + self.global_avgpool = nn.AdaptiveAvgPool2d((1, 1)) + + self.classifier = nn.Linear(output_channels, num_classes) + + def featuremaps(self, x): + x = self.conv1(x) + x = self.maxpool(x) + x = self.stage2(x) + x = self.stage3(x) + x = self.stage4(x) + x = self.conv5(x) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError("Unsupported loss: {}".format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + if model_url is None: + import warnings + warnings.warn( + 'ImageNet pretrained weights are unavailable for this model' + ) + return + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def shufflenet_v2_x0_5(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ShuffleNetV2( + num_classes, loss, [4, 8, 4], [24, 48, 96, 192, 1024], **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['shufflenetv2_x0.5']) + return model + + +def shufflenet_v2_x1_0(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ShuffleNetV2( + num_classes, loss, [4, 8, 4], [24, 116, 232, 464, 1024], **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['shufflenetv2_x1.0']) + return model + + +def shufflenet_v2_x1_5(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ShuffleNetV2( + num_classes, loss, [4, 8, 4], [24, 176, 352, 704, 1024], **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['shufflenetv2_x1.5']) + return model + + +def shufflenet_v2_x2_0(num_classes, loss='softmax', pretrained=True, **kwargs): + model = ShuffleNetV2( + num_classes, loss, [4, 8, 4], [24, 244, 488, 976, 2048], **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['shufflenetv2_x2.0']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py new file mode 100644 index 0000000000..2b9ebb56f2 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/squeezenet.py @@ -0,0 +1,281 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +Code source: https://github.com/pytorch/vision +""" +from __future__ import division, absolute_import +import torch +import torch.nn as nn +import torch.utils.model_zoo as model_zoo + +__all__ = ['squeezenet1_0', 'squeezenet1_1', 'squeezenet1_0_fc512'] + +model_urls = { + 'squeezenet1_0': + 'https://download.pytorch.org/models/squeezenet1_0-a815701f.pth', + 'squeezenet1_1': + 'https://download.pytorch.org/models/squeezenet1_1-f364aa15.pth', +} + + +class Fire(nn.Module): + + def __init__( + self, inplanes, squeeze_planes, expand1x1_planes, expand3x3_planes + ): + super(Fire, self).__init__() + self.inplanes = inplanes + self.squeeze = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1) + self.squeeze_activation = nn.ReLU(inplace=True) + self.expand1x1 = nn.Conv2d( + squeeze_planes, expand1x1_planes, kernel_size=1 + ) + self.expand1x1_activation = nn.ReLU(inplace=True) + self.expand3x3 = nn.Conv2d( + squeeze_planes, expand3x3_planes, kernel_size=3, padding=1 + ) + self.expand3x3_activation = nn.ReLU(inplace=True) + + def forward(self, x): + x = self.squeeze_activation(self.squeeze(x)) + return torch.cat( + [ + self.expand1x1_activation(self.expand1x1(x)), + self.expand3x3_activation(self.expand3x3(x)) + ], 1 + ) + + +class SqueezeNet(nn.Module): + """SqueezeNet. + + Reference: + Iandola et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters + and< 0.5 MB model size. arXiv:1602.07360. + + Public keys: + - ``squeezenet1_0``: SqueezeNet (version=1.0). + - ``squeezenet1_1``: SqueezeNet (version=1.1). + - ``squeezenet1_0_fc512``: SqueezeNet (version=1.0) + FC. + """ + + def __init__( + self, + num_classes, + loss, + version=1.0, + fc_dims=None, + dropout_p=None, + **kwargs + ): + super(SqueezeNet, self).__init__() + self.loss = loss + self.feature_dim = 512 + + if version not in [1.0, 1.1]: + raise ValueError( + 'Unsupported SqueezeNet version {version}:' + '1.0 or 1.1 expected'.format(version=version) + ) + + if version == 1.0: + self.features = nn.Sequential( + nn.Conv2d(3, 96, kernel_size=7, stride=2), + nn.ReLU(inplace=True), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(96, 16, 64, 64), + Fire(128, 16, 64, 64), + Fire(128, 32, 128, 128), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(256, 32, 128, 128), + Fire(256, 48, 192, 192), + Fire(384, 48, 192, 192), + Fire(384, 64, 256, 256), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(512, 64, 256, 256), + ) + else: + self.features = nn.Sequential( + nn.Conv2d(3, 64, kernel_size=3, stride=2), + nn.ReLU(inplace=True), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(64, 16, 64, 64), + Fire(128, 16, 64, 64), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(128, 32, 128, 128), + Fire(256, 32, 128, 128), + nn.MaxPool2d(kernel_size=3, stride=2, ceil_mode=True), + Fire(256, 48, 192, 192), + Fire(384, 48, 192, 192), + Fire(384, 64, 256, 256), + Fire(512, 64, 256, 256), + ) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.fc = self._construct_fc_layer(fc_dims, 512, dropout_p) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def forward(self, x): + f = self.features(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initializes model with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url, map_location=None) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def squeezenet1_0(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SqueezeNet( + num_classes, loss, version=1.0, fc_dims=None, dropout_p=None, **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['squeezenet1_0']) + return model + + +def squeezenet1_0_fc512( + num_classes, loss='softmax', pretrained=True, **kwargs +): + model = SqueezeNet( + num_classes, + loss, + version=1.0, + fc_dims=[512], + dropout_p=None, + **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['squeezenet1_0']) + return model + + +def squeezenet1_1(num_classes, loss='softmax', pretrained=True, **kwargs): + model = SqueezeNet( + num_classes, loss, version=1.1, fc_dims=None, dropout_p=None, **kwargs + ) + if pretrained: + init_pretrained_weights(model, model_urls['squeezenet1_1']) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py new file mode 100644 index 0000000000..ed7039facb --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/models/xception.py @@ -0,0 +1,391 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +import torch.nn as nn +import torch.nn.functional as F +import torch.utils.model_zoo as model_zoo + +__all__ = ['xception'] + +pretrained_settings = { + 'xception': { + 'imagenet': { + 'url': + 'http://data.lip6.fr/cadene/pretrainedmodels/xception-43020ad28.pth', + 'input_space': 'RGB', + 'input_size': [3, 299, 299], + 'input_range': [0, 1], + 'mean': [0.5, 0.5, 0.5], + 'std': [0.5, 0.5, 0.5], + 'num_classes': 1000, + 'scale': + 0.8975 # The resize parameter of the validation transform should be 333, and make sure to center crop at 299x299 + } + } +} + + +class SeparableConv2d(nn.Module): + + def __init__( + self, + in_channels, + out_channels, + kernel_size=1, + stride=1, + padding=0, + dilation=1, + bias=False + ): + super(SeparableConv2d, self).__init__() + + self.conv1 = nn.Conv2d( + in_channels, + in_channels, + kernel_size, + stride, + padding, + dilation, + groups=in_channels, + bias=bias + ) + self.pointwise = nn.Conv2d( + in_channels, out_channels, 1, 1, 0, 1, 1, bias=bias + ) + + def forward(self, x): + x = self.conv1(x) + x = self.pointwise(x) + return x + + +class Block(nn.Module): + + def __init__( + self, + in_filters, + out_filters, + reps, + strides=1, + start_with_relu=True, + grow_first=True + ): + super(Block, self).__init__() + + if out_filters != in_filters or strides != 1: + self.skip = nn.Conv2d( + in_filters, out_filters, 1, stride=strides, bias=False + ) + self.skipbn = nn.BatchNorm2d(out_filters) + else: + self.skip = None + + self.relu = nn.ReLU(inplace=True) + rep = [] + + filters = in_filters + if grow_first: + rep.append(self.relu) + rep.append( + SeparableConv2d( + in_filters, + out_filters, + 3, + stride=1, + padding=1, + bias=False + ) + ) + rep.append(nn.BatchNorm2d(out_filters)) + filters = out_filters + + for i in range(reps - 1): + rep.append(self.relu) + rep.append( + SeparableConv2d( + filters, filters, 3, stride=1, padding=1, bias=False + ) + ) + rep.append(nn.BatchNorm2d(filters)) + + if not grow_first: + rep.append(self.relu) + rep.append( + SeparableConv2d( + in_filters, + out_filters, + 3, + stride=1, + padding=1, + bias=False + ) + ) + rep.append(nn.BatchNorm2d(out_filters)) + + if not start_with_relu: + rep = rep[1:] + else: + rep[0] = nn.ReLU(inplace=False) + + if strides != 1: + rep.append(nn.MaxPool2d(3, strides, 1)) + self.rep = nn.Sequential(*rep) + + def forward(self, inp): + x = self.rep(inp) + + if self.skip is not None: + skip = self.skip(inp) + skip = self.skipbn(skip) + else: + skip = inp + + x += skip + return x + + +class Xception(nn.Module): + """Xception. + + Reference: + Chollet. Xception: Deep Learning with Depthwise + Separable Convolutions. CVPR 2017. + + Public keys: + - ``xception``: Xception. + """ + + def __init__( + self, num_classes, loss, fc_dims=None, dropout_p=None, **kwargs + ): + super(Xception, self).__init__() + self.loss = loss + + self.conv1 = nn.Conv2d(3, 32, 3, 2, 0, bias=False) + self.bn1 = nn.BatchNorm2d(32) + + self.conv2 = nn.Conv2d(32, 64, 3, bias=False) + self.bn2 = nn.BatchNorm2d(64) + + self.block1 = Block( + 64, 128, 2, 2, start_with_relu=False, grow_first=True + ) + self.block2 = Block( + 128, 256, 2, 2, start_with_relu=True, grow_first=True + ) + self.block3 = Block( + 256, 728, 2, 2, start_with_relu=True, grow_first=True + ) + + self.block4 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block5 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block6 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block7 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + + self.block8 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block9 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block10 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + self.block11 = Block( + 728, 728, 3, 1, start_with_relu=True, grow_first=True + ) + + self.block12 = Block( + 728, 1024, 2, 2, start_with_relu=True, grow_first=False + ) + + self.conv3 = SeparableConv2d(1024, 1536, 3, 1, 1) + self.bn3 = nn.BatchNorm2d(1536) + + self.conv4 = SeparableConv2d(1536, 2048, 3, 1, 1) + self.bn4 = nn.BatchNorm2d(2048) + + self.global_avgpool = nn.AdaptiveAvgPool2d(1) + self.feature_dim = 2048 + self.fc = self._construct_fc_layer(fc_dims, 2048, dropout_p) + self.classifier = nn.Linear(self.feature_dim, num_classes) + + self._init_params() + + def _construct_fc_layer(self, fc_dims, input_dim, dropout_p=None): + """Constructs fully connected layer. + + Args: + fc_dims (list or tuple): dimensions of fc layers, if None, no fc layers are constructed + input_dim (int): input dimension + dropout_p (float): dropout probability, if None, dropout is unused + """ + if fc_dims is None: + self.feature_dim = input_dim + return None + + assert isinstance( + fc_dims, (list, tuple) + ), 'fc_dims must be either list or tuple, but got {}'.format( + type(fc_dims) + ) + + layers = [] + for dim in fc_dims: + layers.append(nn.Linear(input_dim, dim)) + layers.append(nn.BatchNorm1d(dim)) + layers.append(nn.ReLU(inplace=True)) + if dropout_p is not None: + layers.append(nn.Dropout(p=dropout_p)) + input_dim = dim + + self.feature_dim = fc_dims[-1] + + return nn.Sequential(*layers) + + def _init_params(self): + for m in self.modules(): + if isinstance(m, nn.Conv2d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu' + ) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm2d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm1d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + if m.bias is not None: + nn.init.constant_(m.bias, 0) + + def featuremaps(self, input): + x = self.conv1(input) + x = self.bn1(x) + x = F.relu(x, inplace=True) + + x = self.conv2(x) + x = self.bn2(x) + x = F.relu(x, inplace=True) + + x = self.block1(x) + x = self.block2(x) + x = self.block3(x) + x = self.block4(x) + x = self.block5(x) + x = self.block6(x) + x = self.block7(x) + x = self.block8(x) + x = self.block9(x) + x = self.block10(x) + x = self.block11(x) + x = self.block12(x) + + x = self.conv3(x) + x = self.bn3(x) + x = F.relu(x, inplace=True) + + x = self.conv4(x) + x = self.bn4(x) + x = F.relu(x, inplace=True) + return x + + def forward(self, x): + f = self.featuremaps(x) + v = self.global_avgpool(f) + v = v.view(v.size(0), -1) + + if self.fc is not None: + v = self.fc(v) + + if not self.training: + return v + + y = self.classifier(v) + + if self.loss == 'softmax': + return y + elif self.loss == 'triplet': + return y, v + else: + raise KeyError('Unsupported loss: {}'.format(self.loss)) + + +def init_pretrained_weights(model, model_url): + """Initialize models with pretrained weights. + + Layers that don't match with pretrained layers in name or size are kept unchanged. + """ + pretrain_dict = model_zoo.load_url(model_url) + model_dict = model.state_dict() + pretrain_dict = { + k: v + for k, v in pretrain_dict.items() + if k in model_dict and model_dict[k].size() == v.size() + } + model_dict.update(pretrain_dict) + model.load_state_dict(model_dict) + + +def xception(num_classes, loss='softmax', pretrained=True, **kwargs): + model = Xception(num_classes, loss, fc_dims=None, dropout_p=None, **kwargs) + if pretrained: + model_url = pretrained_settings['xception']['imagenet']['url'] + init_pretrained_weights(model, model_url) + return model diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py new file mode 100644 index 0000000000..24c0fdc8ef --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/__init__.py @@ -0,0 +1,51 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .optimizer import build_optimizer +from .lr_scheduler import build_lr_scheduler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py new file mode 100644 index 0000000000..da06eb5e8f --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/lr_scheduler.py @@ -0,0 +1,115 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import +import torch + +AVAI_SCH = ['single_step', 'multi_step', 'cosine'] + + +def build_lr_scheduler( + optimizer, lr_scheduler='single_step', stepsize=1, gamma=0.1, max_epoch=1 +): + """A function wrapper for building a learning rate scheduler. + + Args: + optimizer (Optimizer): an Optimizer. + lr_scheduler (str, optional): learning rate scheduler method. Default is single_step. + stepsize (int or list, optional): step size to decay learning rate. When ``lr_scheduler`` + is "single_step", ``stepsize`` should be an integer. When ``lr_scheduler`` is + "multi_step", ``stepsize`` is a list. Default is 1. + gamma (float, optional): decay rate. Default is 0.1. + max_epoch (int, optional): maximum epoch (for cosine annealing). Default is 1. + + Examples:: + >>> # Decay learning rate by every 20 epochs. + >>> scheduler = torchreid.optim.build_lr_scheduler( + >>> optimizer, lr_scheduler='single_step', stepsize=20 + >>> ) + >>> # Decay learning rate at 30, 50 and 55 epochs. + >>> scheduler = torchreid.optim.build_lr_scheduler( + >>> optimizer, lr_scheduler='multi_step', stepsize=[30, 50, 55] + >>> ) + """ + if lr_scheduler not in AVAI_SCH: + raise ValueError( + 'Unsupported scheduler: {}. Must be one of {}'.format( + lr_scheduler, AVAI_SCH + ) + ) + + if lr_scheduler == 'single_step': + if isinstance(stepsize, list): + stepsize = stepsize[-1] + + if not isinstance(stepsize, int): + raise TypeError( + 'For single_step lr_scheduler, stepsize must ' + 'be an integer, but got {}'.format(type(stepsize)) + ) + + scheduler = torch.optim.lr_scheduler.StepLR( + optimizer, step_size=stepsize, gamma=gamma + ) + + elif lr_scheduler == 'multi_step': + if not isinstance(stepsize, list): + raise TypeError( + 'For multi_step lr_scheduler, stepsize must ' + 'be a list, but got {}'.format(type(stepsize)) + ) + + scheduler = torch.optim.lr_scheduler.MultiStepLR( + optimizer, milestones=stepsize, gamma=gamma + ) + + elif lr_scheduler == 'cosine': + scheduler = torch.optim.lr_scheduler.CosineAnnealingLR( + optimizer, float(max_epoch) + ) + + return scheduler diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py new file mode 100644 index 0000000000..73bb6546ea --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/optimizer.py @@ -0,0 +1,218 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import +import warnings +import torch +import torch.nn as nn + +from .radam import RAdam + +import os + +AVAI_OPTIMS = ['adam', 'amsgrad', 'sgd', 'rmsprop', 'radam'] + + +def build_optimizer( + model, + optim='adam', + lr=0.0003, + weight_decay=5e-04, + momentum=0.9, + sgd_dampening=0, + sgd_nesterov=False, + rmsprop_alpha=0.99, + adam_beta1=0.9, + adam_beta2=0.99, + staged_lr=False, + new_layers='', + base_lr_mult=0.1 +): + """A function wrapper for building an optimizer. + + Args: + model (nn.Module): model. + optim (str, optional): optimizer. Default is "adam". + lr (float, optional): learning rate. Default is 0.0003. + weight_decay (float, optional): weight decay (L2 penalty). Default is 5e-04. + momentum (float, optional): momentum factor in sgd. Default is 0.9, + sgd_dampening (float, optional): dampening for momentum. Default is 0. + sgd_nesterov (bool, optional): enables Nesterov momentum. Default is False. + rmsprop_alpha (float, optional): smoothing constant for rmsprop. Default is 0.99. + adam_beta1 (float, optional): beta-1 value in adam. Default is 0.9. + adam_beta2 (float, optional): beta-2 value in adam. Default is 0.99, + staged_lr (bool, optional): uses different learning rates for base and new layers. Base + layers are pretrained layers while new layers are randomly initialized, e.g. the + identity classification layer. Enabling ``staged_lr`` can allow the base layers to + be trained with a smaller learning rate determined by ``base_lr_mult``, while the new + layers will take the ``lr``. Default is False. + new_layers (str or list): attribute names in ``model``. Default is empty. + base_lr_mult (float, optional): learning rate multiplier for base layers. Default is 0.1. + + Examples:: + >>> # A normal optimizer can be built by + >>> optimizer = torchreid.optim.build_optimizer(model, optim='sgd', lr=0.01) + >>> # If you want to use a smaller learning rate for pretrained layers + >>> # and the attribute name for the randomly initialized layer is 'classifier', + >>> # you can do + >>> optimizer = torchreid.optim.build_optimizer( + >>> model, optim='sgd', lr=0.01, staged_lr=True, + >>> new_layers='classifier', base_lr_mult=0.1 + >>> ) + >>> # Now the `classifier` has learning rate 0.01 but the base layers + >>> # have learning rate 0.01 * 0.1. + >>> # new_layers can also take multiple attribute names. Say the new layers + >>> # are 'fc' and 'classifier', you can do + >>> optimizer = torchreid.optim.build_optimizer( + >>> model, optim='sgd', lr=0.01, staged_lr=True, + >>> new_layers=['fc', 'classifier'], base_lr_mult=0.1 + >>> ) + """ + if optim not in AVAI_OPTIMS: + raise ValueError( + 'Unsupported optim: {}. Must be one of {}'.format( + optim, AVAI_OPTIMS + ) + ) + + if not isinstance(model, nn.Module): + raise TypeError( + 'model given to build_optimizer must be an instance of nn.Module' + ) + + if staged_lr: + if isinstance(new_layers, str): + if new_layers is None: + warnings.warn( + 'new_layers is empty, therefore, staged_lr is useless' + ) + new_layers = [new_layers] + + if isinstance(model, nn.DataParallel): + model = model.module + + base_params = [] + base_layers = [] + new_params = [] + + for name, module in model.named_children(): + if name in new_layers: + new_params += [p for p in module.parameters()] + else: + base_params += [p for p in module.parameters()] + base_layers.append(name) + + param_groups = [ + { + 'params': base_params, + 'lr': lr * base_lr_mult + }, + { + 'params': new_params + }, + ] + + else: + param_groups = model.parameters() + + if optim == 'adam': + optimizer = torch.optim.Adam( + param_groups, + lr=lr, + weight_decay=weight_decay, + betas=(adam_beta1, adam_beta2), + ) + + elif optim == 'amsgrad': + optimizer = torch.optim.Adam( + param_groups, + lr=lr, + weight_decay=weight_decay, + betas=(adam_beta1, adam_beta2), + amsgrad=True, + ) + + elif optim == 'sgd': + if os.environ['device'] == "gpu": + optimizer = torch.optim.SGD( + param_groups, + lr=lr, + momentum=momentum, + weight_decay=weight_decay, + dampening=sgd_dampening, + nesterov=sgd_nesterov, + ) + elif os.environ['device'] == "npu": + from apex.optimizers import NpuFusedSGD + optimizer = NpuFusedSGD( + param_groups, + lr=lr, + momentum=momentum, + weight_decay=weight_decay, + dampening=sgd_dampening, + nesterov=sgd_nesterov, + ) + + + elif optim == 'rmsprop': + optimizer = torch.optim.RMSprop( + param_groups, + lr=lr, + momentum=momentum, + weight_decay=weight_decay, + alpha=rmsprop_alpha, + ) + + elif optim == 'radam': + optimizer = RAdam( + param_groups, + lr=lr, + weight_decay=weight_decay, + betas=(adam_beta1, adam_beta2) + ) + + return optimizer diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py new file mode 100644 index 0000000000..79d55b4ac8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/optim/radam.py @@ -0,0 +1,376 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. + +Imported from: https://github.com/LiyuanLucasLiu/RAdam + +Paper: https://arxiv.org/abs/1908.03265 + +@article{liu2019radam, + title={On the Variance of the Adaptive Learning Rate and Beyond}, + author={Liu, Liyuan and Jiang, Haoming and He, Pengcheng and Chen, Weizhu and Liu, Xiaodong and Gao, Jianfeng and Han, Jiawei}, + journal={arXiv preprint arXiv:1908.03265}, + year={2019} +} +""" +from __future__ import print_function, absolute_import +import math +import torch +from torch.optim.optimizer import Optimizer + + +class RAdam(Optimizer): + + def __init__( + self, + params, + lr=1e-3, + betas=(0.9, 0.999), + eps=1e-8, + weight_decay=0, + degenerated_to_sgd=True + ): + if not 0.0 <= lr: + raise ValueError("Invalid learning rate: {}".format(lr)) + if not 0.0 <= eps: + raise ValueError("Invalid epsilon value: {}".format(eps)) + if not 0.0 <= betas[0] < 1.0: + raise ValueError( + "Invalid beta parameter at index 0: {}".format(betas[0]) + ) + if not 0.0 <= betas[1] < 1.0: + raise ValueError( + "Invalid beta parameter at index 1: {}".format(betas[1]) + ) + + self.degenerated_to_sgd = degenerated_to_sgd + defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay) + self.buffer = [[None, None, None] for ind in range(10)] + super(RAdam, self).__init__(params, defaults) + + def __setstate__(self, state): + super(RAdam, self).__setstate__(state) + + def step(self, closure=None): + + loss = None + if closure is not None: + loss = closure() + + for group in self.param_groups: + + for p in group['params']: + if p.grad is None: + continue + grad = p.grad.data.float() + if grad.is_sparse: + raise RuntimeError( + 'RAdam does not support sparse gradients' + ) + + p_data_fp32 = p.data.float() + + state = self.state[p] + + if len(state) == 0: + state['step'] = 0 + state['exp_avg'] = torch.zeros_like(p_data_fp32) + state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) + else: + state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) + state['exp_avg_sq'] = state['exp_avg_sq'].type_as( + p_data_fp32 + ) + + exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] + beta1, beta2 = group['betas'] + + exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) + exp_avg.mul_(beta1).add_(1 - beta1, grad) + + state['step'] += 1 + buffered = self.buffer[int(state['step'] % 10)] + if state['step'] == buffered[0]: + N_sma, step_size = buffered[1], buffered[2] + else: + buffered[0] = state['step'] + beta2_t = beta2**state['step'] + N_sma_max = 2 / (1-beta2) - 1 + N_sma = N_sma_max - 2 * state['step' + ] * beta2_t / (1-beta2_t) + buffered[1] = N_sma + + # more conservative since it's an approximated value + if N_sma >= 5: + step_size = math.sqrt( + (1-beta2_t) * (N_sma-4) / (N_sma_max-4) * + (N_sma-2) / N_sma * N_sma_max / (N_sma_max-2) + ) / (1 - beta1**state['step']) + elif self.degenerated_to_sgd: + step_size = 1.0 / (1 - beta1**state['step']) + else: + step_size = -1 + buffered[2] = step_size + + # more conservative since it's an approximated value + if N_sma >= 5: + if group['weight_decay'] != 0: + p_data_fp32.add_( + -group['weight_decay'] * group['lr'], p_data_fp32 + ) + denom = exp_avg_sq.sqrt().add_(group['eps']) + p_data_fp32.addcdiv_( + -step_size * group['lr'], exp_avg, denom + ) + p.data.copy_(p_data_fp32) + elif step_size > 0: + if group['weight_decay'] != 0: + p_data_fp32.add_( + -group['weight_decay'] * group['lr'], p_data_fp32 + ) + p_data_fp32.add_(-step_size * group['lr'], exp_avg) + p.data.copy_(p_data_fp32) + + return loss + + +class PlainRAdam(Optimizer): + + def __init__( + self, + params, + lr=1e-3, + betas=(0.9, 0.999), + eps=1e-8, + weight_decay=0, + degenerated_to_sgd=True + ): + if not 0.0 <= lr: + raise ValueError("Invalid learning rate: {}".format(lr)) + if not 0.0 <= eps: + raise ValueError("Invalid epsilon value: {}".format(eps)) + if not 0.0 <= betas[0] < 1.0: + raise ValueError( + "Invalid beta parameter at index 0: {}".format(betas[0]) + ) + if not 0.0 <= betas[1] < 1.0: + raise ValueError( + "Invalid beta parameter at index 1: {}".format(betas[1]) + ) + + self.degenerated_to_sgd = degenerated_to_sgd + defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay) + + super(PlainRAdam, self).__init__(params, defaults) + + def __setstate__(self, state): + super(PlainRAdam, self).__setstate__(state) + + def step(self, closure=None): + + loss = None + if closure is not None: + loss = closure() + + for group in self.param_groups: + + for p in group['params']: + if p.grad is None: + continue + grad = p.grad.data.float() + if grad.is_sparse: + raise RuntimeError( + 'RAdam does not support sparse gradients' + ) + + p_data_fp32 = p.data.float() + + state = self.state[p] + + if len(state) == 0: + state['step'] = 0 + state['exp_avg'] = torch.zeros_like(p_data_fp32) + state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) + else: + state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) + state['exp_avg_sq'] = state['exp_avg_sq'].type_as( + p_data_fp32 + ) + + exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] + beta1, beta2 = group['betas'] + + exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) + exp_avg.mul_(beta1).add_(1 - beta1, grad) + + state['step'] += 1 + beta2_t = beta2**state['step'] + N_sma_max = 2 / (1-beta2) - 1 + N_sma = N_sma_max - 2 * state['step'] * beta2_t / (1-beta2_t) + + # more conservative since it's an approximated value + if N_sma >= 5: + if group['weight_decay'] != 0: + p_data_fp32.add_( + -group['weight_decay'] * group['lr'], p_data_fp32 + ) + step_size = group['lr'] * math.sqrt( + (1-beta2_t) * (N_sma-4) / (N_sma_max-4) * + (N_sma-2) / N_sma * N_sma_max / (N_sma_max-2) + ) / (1 - beta1**state['step']) + denom = exp_avg_sq.sqrt().add_(group['eps']) + p_data_fp32.addcdiv_(-step_size, exp_avg, denom) + p.data.copy_(p_data_fp32) + elif self.degenerated_to_sgd: + if group['weight_decay'] != 0: + p_data_fp32.add_( + -group['weight_decay'] * group['lr'], p_data_fp32 + ) + step_size = group['lr'] / (1 - beta1**state['step']) + p_data_fp32.add_(-step_size, exp_avg) + p.data.copy_(p_data_fp32) + + return loss + + +class AdamW(Optimizer): + + def __init__( + self, + params, + lr=1e-3, + betas=(0.9, 0.999), + eps=1e-8, + weight_decay=0, + warmup=0 + ): + if not 0.0 <= lr: + raise ValueError("Invalid learning rate: {}".format(lr)) + if not 0.0 <= eps: + raise ValueError("Invalid epsilon value: {}".format(eps)) + if not 0.0 <= betas[0] < 1.0: + raise ValueError( + "Invalid beta parameter at index 0: {}".format(betas[0]) + ) + if not 0.0 <= betas[1] < 1.0: + raise ValueError( + "Invalid beta parameter at index 1: {}".format(betas[1]) + ) + + defaults = dict( + lr=lr, + betas=betas, + eps=eps, + weight_decay=weight_decay, + warmup=warmup + ) + super(AdamW, self).__init__(params, defaults) + + def __setstate__(self, state): + super(AdamW, self).__setstate__(state) + + def step(self, closure=None): + loss = None + if closure is not None: + loss = closure() + + for group in self.param_groups: + + for p in group['params']: + if p.grad is None: + continue + grad = p.grad.data.float() + if grad.is_sparse: + raise RuntimeError( + 'Adam does not support sparse gradients, please consider SparseAdam instead' + ) + + p_data_fp32 = p.data.float() + + state = self.state[p] + + if len(state) == 0: + state['step'] = 0 + state['exp_avg'] = torch.zeros_like(p_data_fp32) + state['exp_avg_sq'] = torch.zeros_like(p_data_fp32) + else: + state['exp_avg'] = state['exp_avg'].type_as(p_data_fp32) + state['exp_avg_sq'] = state['exp_avg_sq'].type_as( + p_data_fp32 + ) + + exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq'] + beta1, beta2 = group['betas'] + + state['step'] += 1 + + exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad) + exp_avg.mul_(beta1).add_(1 - beta1, grad) + + denom = exp_avg_sq.sqrt().add_(group['eps']) + bias_correction1 = 1 - beta1**state['step'] + bias_correction2 = 1 - beta2**state['step'] + + if group['warmup'] > state['step']: + scheduled_lr = 1e-8 + state['step'] * group['lr'] / group[ + 'warmup'] + else: + scheduled_lr = group['lr'] + + step_size = scheduled_lr * math.sqrt( + bias_correction2 + ) / bias_correction1 + + if group['weight_decay'] != 0: + p_data_fp32.add_( + -group['weight_decay'] * scheduled_lr, p_data_fp32 + ) + + p_data_fp32.addcdiv_(-step_size, exp_avg, denom) + + p.data.copy_(p_data_fp32) + + return loss diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py new file mode 100644 index 0000000000..8f39046e67 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/__init__.py @@ -0,0 +1,57 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import + +from .tools import * +from .rerank import re_ranking +from .loggers import * +from .avgmeter import * +from .reidtools import * +from .torchtools import * +from .model_complexity import compute_model_complexity +from .feature_extractor import FeatureExtractor diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py new file mode 100644 index 0000000000..82ef00ed66 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/avgmeter.py @@ -0,0 +1,120 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, absolute_import +from collections import defaultdict +import torch + +__all__ = ['AverageMeter', 'MetricMeter'] + + +class AverageMeter(object): + """Computes and stores the average and current value. + + Examples:: + >>> # Initialize a meter to record loss + >>> losses = AverageMeter() + >>> # Update meter after every minibatch update + >>> losses.update(loss_value, batch_size) + """ + + def __init__(self): + self.reset() + + def reset(self): + self.val = 0 + self.avg = 0 + self.sum = 0 + self.count = 0 + + def update(self, val, n=1): + self.val = val + self.sum += val * n + self.count += n + self.avg = self.sum / self.count + + +class MetricMeter(object): + """A collection of metrics. + + Source: https://github.com/KaiyangZhou/Dassl.pytorch + + Examples:: + >>> # 1. Create an instance of MetricMeter + >>> metric = MetricMeter() + >>> # 2. Update using a dictionary as input + >>> input_dict = {'loss_1': value_1, 'loss_2': value_2} + >>> metric.update(input_dict) + >>> # 3. Convert to string and print + >>> print(str(metric)) + """ + + def __init__(self, delimiter='\t'): + self.meters = defaultdict(AverageMeter) + self.delimiter = delimiter + + def update(self, input_dict): + if input_dict is None: + return + + if not isinstance(input_dict, dict): + raise TypeError( + 'Input to MetricMeter.update() must be a dictionary' + ) + + for k, v in input_dict.items(): + if isinstance(v, torch.Tensor): + v = v.item() + self.meters[k].update(v) + + def __str__(self): + output_str = [] + for name, meter in self.meters.items(): + output_str.append( + '{} {:.4f} ({:.4f})'.format(name, meter.val, meter.avg) + ) + return self.delimiter.join(output_str) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py new file mode 100644 index 0000000000..b25cc9959d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/feature_extractor.py @@ -0,0 +1,199 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import +import numpy as np +import torch +import torchvision.transforms as T +from PIL import Image + +from torchreid.utils import ( + check_isfile, load_pretrained_weights, compute_model_complexity +) +from torchreid.models import build_model + + +class FeatureExtractor(object): + """A simple API for feature extraction. + + FeatureExtractor can be used like a python function, which + accepts input of the following types: + - a list of strings (image paths) + - a list of numpy.ndarray each with shape (H, W, C) + - a single string (image path) + - a single numpy.ndarray with shape (H, W, C) + - a torch.Tensor with shape (B, C, H, W) or (C, H, W) + + Returned is a torch tensor with shape (B, D) where D is the + feature dimension. + + Args: + model_name (str): model name. + model_path (str): path to model weights. + image_size (sequence or int): image height and width. + pixel_mean (list): pixel mean for normalization. + pixel_std (list): pixel std for normalization. + pixel_norm (bool): whether to normalize pixels. + device (str): 'cpu' or 'cuda' (could be specific gpu devices). + verbose (bool): show model details. + + Examples:: + + from torchreid.utils import FeatureExtractor + + extractor = FeatureExtractor( + model_name='osnet_x1_0', + model_path='a/b/c/model.pth.tar', + device='cuda' + ) + + image_list = [ + 'a/b/c/image001.jpg', + 'a/b/c/image002.jpg', + 'a/b/c/image003.jpg', + 'a/b/c/image004.jpg', + 'a/b/c/image005.jpg' + ] + + features = extractor(image_list) + print(features.shape) # output (5, 512) + """ + + def __init__( + self, + model_name='', + model_path='', + image_size=(256, 128), + pixel_mean=[0.485, 0.456, 0.406], + pixel_std=[0.229, 0.224, 0.225], + pixel_norm=True, + device='cuda', + verbose=True + ): + # Build model + model = build_model( + model_name, + num_classes=1, + pretrained=True, + use_gpu=device.startswith('cuda') + ) + model.eval() + + if verbose: + num_params, flops = compute_model_complexity( + model, (1, 3, image_size[0], image_size[1]) + ) + print('Model: {}'.format(model_name)) + print('- params: {:,}'.format(num_params)) + print('- flops: {:,}'.format(flops)) + + if model_path and check_isfile(model_path): + load_pretrained_weights(model, model_path) + + # Build transform functions + transforms = [] + transforms += [T.Resize(image_size)] + transforms += [T.ToTensor()] + if pixel_norm: + transforms += [T.Normalize(mean=pixel_mean, std=pixel_std)] + preprocess = T.Compose(transforms) + + to_pil = T.ToPILImage() + + device = torch.device(device) + model.to(device) + + # Class attributes + self.model = model + self.preprocess = preprocess + self.to_pil = to_pil + self.device = device + + def __call__(self, input): + if isinstance(input, list): + images = [] + + for element in input: + if isinstance(element, str): + image = Image.open(element).convert('RGB') + + elif isinstance(element, np.ndarray): + image = self.to_pil(element) + + else: + raise TypeError( + 'Type of each element must belong to [str | numpy.ndarray]' + ) + + image = self.preprocess(image) + images.append(image) + + images = torch.stack(images, dim=0) + images = images.to(self.device) + + elif isinstance(input, str): + image = Image.open(input).convert('RGB') + image = self.preprocess(image) + images = image.unsqueeze(0).to(self.device) + + elif isinstance(input, np.ndarray): + image = self.to_pil(input) + image = self.preprocess(image) + images = image.unsqueeze(0).to(self.device) + + elif isinstance(input, torch.Tensor): + if input.dim() == 3: + input = input.unsqueeze(0) + images = input.to(self.device) + + else: + raise NotImplementedError + + with torch.no_grad(): + features = self.model(images) + + return features diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py new file mode 100644 index 0000000000..f61ad5cd91 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/loggers.py @@ -0,0 +1,193 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import absolute_import +import os +import sys +import os.path as osp + +from .tools import mkdir_if_missing + +__all__ = ['Logger', 'RankLogger'] + + +class Logger(object): + """Writes console output to external text file. + + Imported from ``_ + + Args: + fpath (str): directory to save logging file. + + Examples:: + >>> import sys + >>> import os + >>> import os.path as osp + >>> from torchreid.utils import Logger + >>> save_dir = 'log/resnet50-softmax-market1501' + >>> log_name = 'train.log' + >>> sys.stdout = Logger(osp.join(args.save_dir, log_name)) + """ + + def __init__(self, fpath=None): + self.console = sys.stdout + self.file = None + if fpath is not None: + mkdir_if_missing(osp.dirname(fpath)) + self.file = open(fpath, 'w') + + def __del__(self): + self.close() + + def __enter__(self): + pass + + def __exit__(self, *args): + self.close() + + def write(self, msg): + self.console.write(msg) + if self.file is not None: + self.file.write(msg) + + def flush(self): + self.console.flush() + if self.file is not None: + self.file.flush() + os.fsync(self.file.fileno()) + + def close(self): + self.console.close() + if self.file is not None: + self.file.close() + + +class RankLogger(object): + """Records the rank1 matching accuracy obtained for each + test dataset at specified evaluation steps and provides a function + to show the summarized results, which are convenient for analysis. + + Args: + sources (str or list): source dataset name(s). + targets (str or list): target dataset name(s). + + Examples:: + >>> from torchreid.utils import RankLogger + >>> s = 'market1501' + >>> t = 'market1501' + >>> ranklogger = RankLogger(s, t) + >>> ranklogger.write(t, 10, 0.5) + >>> ranklogger.write(t, 20, 0.7) + >>> ranklogger.write(t, 30, 0.9) + >>> ranklogger.show_summary() + >>> # You will see: + >>> # => Show performance summary + >>> # market1501 (source) + >>> # - epoch 10 rank1 50.0% + >>> # - epoch 20 rank1 70.0% + >>> # - epoch 30 rank1 90.0% + >>> # If there are multiple test datasets + >>> t = ['market1501', 'dukemtmcreid'] + >>> ranklogger = RankLogger(s, t) + >>> ranklogger.write(t[0], 10, 0.5) + >>> ranklogger.write(t[0], 20, 0.7) + >>> ranklogger.write(t[0], 30, 0.9) + >>> ranklogger.write(t[1], 10, 0.1) + >>> ranklogger.write(t[1], 20, 0.2) + >>> ranklogger.write(t[1], 30, 0.3) + >>> ranklogger.show_summary() + >>> # You can see: + >>> # => Show performance summary + >>> # market1501 (source) + >>> # - epoch 10 rank1 50.0% + >>> # - epoch 20 rank1 70.0% + >>> # - epoch 30 rank1 90.0% + >>> # dukemtmcreid (target) + >>> # - epoch 10 rank1 10.0% + >>> # - epoch 20 rank1 20.0% + >>> # - epoch 30 rank1 30.0% + """ + + def __init__(self, sources, targets): + self.sources = sources + self.targets = targets + + if isinstance(self.sources, str): + self.sources = [self.sources] + + if isinstance(self.targets, str): + self.targets = [self.targets] + + self.logger = { + name: { + 'epoch': [], + 'rank1': [] + } + for name in self.targets + } + + def write(self, name, epoch, rank1): + """Writes result. + + Args: + name (str): dataset name. + epoch (int): current epoch. + rank1 (float): rank1 result. + """ + self.logger[name]['epoch'].append(epoch) + self.logger[name]['rank1'].append(rank1) + + def show_summary(self): + """Shows saved results.""" + print('=> Show performance summary') + for name in self.targets: + from_where = 'source' if name in self.sources else 'target' + print('{} ({})'.format(name, from_where)) + for epoch, rank1 in zip( + self.logger[name]['epoch'], self.logger[name]['rank1'] + ): + print('- epoch {}\t rank1 {:.1%}'.format(epoch, rank1)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py new file mode 100644 index 0000000000..af673038e8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/model_complexity.py @@ -0,0 +1,410 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import math +import numpy as np +from itertools import repeat +from collections import namedtuple, defaultdict +import torch + +__all__ = ['compute_model_complexity'] +""" +Utility +""" + + +def _ntuple(n): + + def parse(x): + if isinstance(x, int): + return tuple(repeat(x, n)) + return x + + return parse + + +_single = _ntuple(1) +_pair = _ntuple(2) +_triple = _ntuple(3) +""" +Convolution +""" + + +def hook_convNd(m, x, y): + k = torch.prod(torch.Tensor(m.kernel_size)).item() + cin = m.in_channels + flops_per_ele = k * cin # + (k*cin-1) + if m.bias is not None: + flops_per_ele += 1 + flops = flops_per_ele * y.numel() / m.groups + return int(flops) + + +""" +Pooling +""" + + +def hook_maxpool1d(m, x, y): + flops_per_ele = m.kernel_size - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_maxpool2d(m, x, y): + k = _pair(m.kernel_size) + k = torch.prod(torch.Tensor(k)).item() + # ops: compare + flops_per_ele = k - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_maxpool3d(m, x, y): + k = _triple(m.kernel_size) + k = torch.prod(torch.Tensor(k)).item() + flops_per_ele = k - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_avgpool1d(m, x, y): + flops_per_ele = m.kernel_size + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_avgpool2d(m, x, y): + k = _pair(m.kernel_size) + k = torch.prod(torch.Tensor(k)).item() + flops_per_ele = k + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_avgpool3d(m, x, y): + k = _triple(m.kernel_size) + k = torch.prod(torch.Tensor(k)).item() + flops_per_ele = k + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapmaxpool1d(m, x, y): + x = x[0] + out_size = m.output_size + k = math.ceil(x.size(2) / out_size) + flops_per_ele = k - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapmaxpool2d(m, x, y): + x = x[0] + out_size = _pair(m.output_size) + k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) + k = torch.prod(torch.ceil(k)).item() + flops_per_ele = k - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapmaxpool3d(m, x, y): + x = x[0] + out_size = _triple(m.output_size) + k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) + k = torch.prod(torch.ceil(k)).item() + flops_per_ele = k - 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapavgpool1d(m, x, y): + x = x[0] + out_size = m.output_size + k = math.ceil(x.size(2) / out_size) + flops_per_ele = k + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapavgpool2d(m, x, y): + x = x[0] + out_size = _pair(m.output_size) + k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) + k = torch.prod(torch.ceil(k)).item() + flops_per_ele = k + flops = flops_per_ele * y.numel() + return int(flops) + + +def hook_adapavgpool3d(m, x, y): + x = x[0] + out_size = _triple(m.output_size) + k = torch.Tensor(list(x.size()[2:])) / torch.Tensor(out_size) + k = torch.prod(torch.ceil(k)).item() + flops_per_ele = k + flops = flops_per_ele * y.numel() + return int(flops) + + +""" +Non-linear activations +""" + + +def hook_relu(m, x, y): + # eq: max(0, x) + num_ele = y.numel() + return int(num_ele) + + +def hook_leakyrelu(m, x, y): + # eq: max(0, x) + negative_slope*min(0, x) + num_ele = y.numel() + flops = 3 * num_ele + return int(flops) + + +""" +Normalization +""" + + +def hook_batchnormNd(m, x, y): + num_ele = y.numel() + flops = 2 * num_ele # mean and std + if m.affine: + flops += 2 * num_ele # gamma and beta + return int(flops) + + +def hook_instancenormNd(m, x, y): + return hook_batchnormNd(m, x, y) + + +def hook_groupnorm(m, x, y): + return hook_batchnormNd(m, x, y) + + +def hook_layernorm(m, x, y): + num_ele = y.numel() + flops = 2 * num_ele # mean and std + if m.elementwise_affine: + flops += 2 * num_ele # gamma and beta + return int(flops) + + +""" +Linear +""" + + +def hook_linear(m, x, y): + flops_per_ele = m.in_features # + (m.in_features-1) + if m.bias is not None: + flops_per_ele += 1 + flops = flops_per_ele * y.numel() + return int(flops) + + +__generic_flops_counter = { + # Convolution + 'Conv1d': hook_convNd, + 'Conv2d': hook_convNd, + 'Conv3d': hook_convNd, + # Pooling + 'MaxPool1d': hook_maxpool1d, + 'MaxPool2d': hook_maxpool2d, + 'MaxPool3d': hook_maxpool3d, + 'AvgPool1d': hook_avgpool1d, + 'AvgPool2d': hook_avgpool2d, + 'AvgPool3d': hook_avgpool3d, + 'AdaptiveMaxPool1d': hook_adapmaxpool1d, + 'AdaptiveMaxPool2d': hook_adapmaxpool2d, + 'AdaptiveMaxPool3d': hook_adapmaxpool3d, + 'AdaptiveAvgPool1d': hook_adapavgpool1d, + 'AdaptiveAvgPool2d': hook_adapavgpool2d, + 'AdaptiveAvgPool3d': hook_adapavgpool3d, + # Non-linear activations + 'ReLU': hook_relu, + 'ReLU6': hook_relu, + 'LeakyReLU': hook_leakyrelu, + # Normalization + 'BatchNorm1d': hook_batchnormNd, + 'BatchNorm2d': hook_batchnormNd, + 'BatchNorm3d': hook_batchnormNd, + 'InstanceNorm1d': hook_instancenormNd, + 'InstanceNorm2d': hook_instancenormNd, + 'InstanceNorm3d': hook_instancenormNd, + 'GroupNorm': hook_groupnorm, + 'LayerNorm': hook_layernorm, + # Linear + 'Linear': hook_linear, +} + +__conv_linear_flops_counter = { + # Convolution + 'Conv1d': hook_convNd, + 'Conv2d': hook_convNd, + 'Conv3d': hook_convNd, + # Linear + 'Linear': hook_linear, +} + + +def _get_flops_counter(only_conv_linear): + if only_conv_linear: + return __conv_linear_flops_counter + return __generic_flops_counter + + +def compute_model_complexity( + model, input_size, verbose=False, only_conv_linear=True +): + """Returns number of parameters and FLOPs. + + .. note:: + (1) this function only provides an estimate of the theoretical time complexity + rather than the actual running time which depends on implementations and hardware, + and (2) the FLOPs is only counted for layers that are used at test time. This means + that redundant layers such as person ID classification layer will be ignored as it + is discarded when doing feature extraction. Note that the inference graph depends on + how you construct the computations in ``forward()``. + + Args: + model (nn.Module): network model. + input_size (tuple): input size, e.g. (1, 3, 256, 128). + verbose (bool, optional): shows detailed complexity of + each module. Default is False. + only_conv_linear (bool, optional): only considers convolution + and linear layers when counting flops. Default is True. + If set to False, flops of all layers will be counted. + + Examples:: + >>> from torchreid import models, utils + >>> model = models.build_model(name='resnet50', num_classes=1000) + >>> num_params, flops = utils.compute_model_complexity(model, (1, 3, 256, 128), verbose=True) + """ + registered_handles = [] + layer_list = [] + layer = namedtuple('layer', ['class_name', 'params', 'flops']) + + def _add_hooks(m): + + def _has_submodule(m): + return len(list(m.children())) > 0 + + def _hook(m, x, y): + params = sum(p.numel() for p in m.parameters()) + class_name = str(m.__class__.__name__) + flops_counter = _get_flops_counter(only_conv_linear) + if class_name in flops_counter: + flops = flops_counter[class_name](m, x, y) + else: + flops = 0 + layer_list.append( + layer(class_name=class_name, params=params, flops=flops) + ) + + # only consider the very basic nn layer + if _has_submodule(m): + return + + handle = m.register_forward_hook(_hook) + registered_handles.append(handle) + + default_train_mode = model.training + + model.eval().apply(_add_hooks) + input = torch.rand(input_size) + if next(model.parameters()).is_cuda: + input = input.cuda() + model(input) # forward + + for handle in registered_handles: + handle.remove() + + model.train(default_train_mode) + + if verbose: + per_module_params = defaultdict(list) + per_module_flops = defaultdict(list) + + total_params, total_flops = 0, 0 + + for layer in layer_list: + total_params += layer.params + total_flops += layer.flops + if verbose: + per_module_params[layer.class_name].append(layer.params) + per_module_flops[layer.class_name].append(layer.flops) + + if verbose: + num_udscore = 55 + print(' {}'.format('-' * num_udscore)) + print(' Model complexity with input size {}'.format(input_size)) + print(' {}'.format('-' * num_udscore)) + for class_name in per_module_params: + params = int(np.sum(per_module_params[class_name])) + flops = int(np.sum(per_module_flops[class_name])) + print( + ' {} (params={:,}, flops={:,})'.format( + class_name, params, flops + ) + ) + print(' {}'.format('-' * num_udscore)) + print( + ' Total (params={:,}, flops={:,})'.format( + total_params, total_flops + ) + ) + print(' {}'.format('-' * num_udscore)) + + return total_params, total_flops diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py new file mode 100644 index 0000000000..4cf1a7ae74 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/reidtools.py @@ -0,0 +1,201 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import print_function, absolute_import +import numpy as np +import shutil +import os.path as osp +import cv2 + +from .tools import mkdir_if_missing + +__all__ = ['visualize_ranked_results'] + +GRID_SPACING = 10 +QUERY_EXTRA_SPACING = 90 +BW = 5 # border width +GREEN = (0, 255, 0) +RED = (0, 0, 255) + + +def visualize_ranked_results( + distmat, dataset, data_type, width=128, height=256, save_dir='', topk=10 +): + """Visualizes ranked results. + + Supports both image-reid and video-reid. + + For image-reid, ranks will be plotted in a single figure. For video-reid, ranks will be + saved in folders each containing a tracklet. + + Args: + distmat (numpy.ndarray): distance matrix of shape (num_query, num_gallery). + dataset (tuple): a 2-tuple containing (query, gallery), each of which contains + tuples of (img_path(s), pid, camid, dsetid). + data_type (str): "image" or "video". + width (int, optional): resized image width. Default is 128. + height (int, optional): resized image height. Default is 256. + save_dir (str): directory to save output images. + topk (int, optional): denoting top-k images in the rank list to be visualized. + Default is 10. + """ + num_q, num_g = distmat.shape + mkdir_if_missing(save_dir) + + print('# query: {}\n# gallery {}'.format(num_q, num_g)) + print('Visualizing top-{} ranks ...'.format(topk)) + + query, gallery = dataset + assert num_q == len(query) + assert num_g == len(gallery) + + indices = np.argsort(distmat, axis=1) + + def _cp_img_to(src, dst, rank, prefix, matched=False): + """ + Args: + src: image path or tuple (for vidreid) + dst: target directory + rank: int, denoting ranked position, starting from 1 + prefix: string + matched: bool + """ + if isinstance(src, (tuple, list)): + if prefix == 'gallery': + suffix = 'TRUE' if matched else 'FALSE' + dst = osp.join( + dst, prefix + '_top' + str(rank).zfill(3) + ) + '_' + suffix + else: + dst = osp.join(dst, prefix + '_top' + str(rank).zfill(3)) + mkdir_if_missing(dst) + for img_path in src: + shutil.copy(img_path, dst) + else: + dst = osp.join( + dst, prefix + '_top' + str(rank).zfill(3) + '_name_' + + osp.basename(src) + ) + shutil.copy(src, dst) + + for q_idx in range(num_q): + qimg_path, qpid, qcamid = query[q_idx][:3] + qimg_path_name = qimg_path[0] if isinstance( + qimg_path, (tuple, list) + ) else qimg_path + + if data_type == 'image': + qimg = cv2.imread(qimg_path) + qimg = cv2.resize(qimg, (width, height)) + qimg = cv2.copyMakeBorder( + qimg, BW, BW, BW, BW, cv2.BORDER_CONSTANT, value=(0, 0, 0) + ) + # resize twice to ensure that the border width is consistent across images + qimg = cv2.resize(qimg, (width, height)) + num_cols = topk + 1 + grid_img = 255 * np.ones( + ( + height, + num_cols*width + topk*GRID_SPACING + QUERY_EXTRA_SPACING, 3 + ), + dtype=np.uint8 + ) + grid_img[:, :width, :] = qimg + else: + qdir = osp.join( + save_dir, osp.basename(osp.splitext(qimg_path_name)[0]) + ) + mkdir_if_missing(qdir) + _cp_img_to(qimg_path, qdir, rank=0, prefix='query') + + rank_idx = 1 + for g_idx in indices[q_idx, :]: + gimg_path, gpid, gcamid = gallery[g_idx][:3] + invalid = (qpid == gpid) & (qcamid == gcamid) + + if not invalid: + matched = gpid == qpid + if data_type == 'image': + border_color = GREEN if matched else RED + gimg = cv2.imread(gimg_path) + gimg = cv2.resize(gimg, (width, height)) + gimg = cv2.copyMakeBorder( + gimg, + BW, + BW, + BW, + BW, + cv2.BORDER_CONSTANT, + value=border_color + ) + gimg = cv2.resize(gimg, (width, height)) + start = rank_idx*width + rank_idx*GRID_SPACING + QUERY_EXTRA_SPACING + end = ( + rank_idx+1 + ) * width + rank_idx*GRID_SPACING + QUERY_EXTRA_SPACING + grid_img[:, start:end, :] = gimg + else: + _cp_img_to( + gimg_path, + qdir, + rank=rank_idx, + prefix='gallery', + matched=matched + ) + + rank_idx += 1 + if rank_idx > topk: + break + + if data_type == 'image': + imname = osp.basename(osp.splitext(qimg_path_name)[0]) + cv2.imwrite(osp.join(save_dir, imname + '.jpg'), grid_img) + + if (q_idx+1) % 100 == 0: + print('- done {}/{}'.format(q_idx + 1, num_q)) + + print('Done. Images have been saved to "{}" ...'.format(save_dir)) diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py new file mode 100644 index 0000000000..f3e0ab8faf --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/rerank.py @@ -0,0 +1,157 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. + +Source: https://github.com/zhunzhong07/person-re-ranking + +Created on Mon Jun 26 14:46:56 2017 +@author: luohao +Modified by Houjing Huang, 2017-12-22. +- This version accepts distance matrix instead of raw features. +- The difference of `/` division between python 2 and 3 is handled. +- numpy.float16 is replaced by numpy.float32 for numerical precision. + +CVPR2017 paper:Zhong Z, Zheng L, Cao D, et al. Re-ranking Person Re-identification with k-reciprocal Encoding[J]. 2017. +url:http://openaccess.thecvf.com/content_cvpr_2017/papers/Zhong_Re-Ranking_Person_Re-Identification_CVPR_2017_paper.pdf +Matlab version: https://github.com/zhunzhong07/person-re-ranking + +API +q_g_dist: query-gallery distance matrix, numpy array, shape [num_query, num_gallery] +q_q_dist: query-query distance matrix, numpy array, shape [num_query, num_query] +g_g_dist: gallery-gallery distance matrix, numpy array, shape [num_gallery, num_gallery] +k1, k2, lambda_value: parameters, the original paper is (k1=20, k2=6, lambda_value=0.3) +Returns: + final_dist: re-ranked distance, numpy array, shape [num_query, num_gallery] +""" +from __future__ import division, print_function, absolute_import +import numpy as np + +__all__ = ['re_ranking'] + + +def re_ranking(q_g_dist, q_q_dist, g_g_dist, k1=20, k2=6, lambda_value=0.3): + + # The following naming, e.g. gallery_num, is different from outer scope. + # Don't care about it. + + original_dist = np.concatenate( + [ + np.concatenate([q_q_dist, q_g_dist], axis=1), + np.concatenate([q_g_dist.T, g_g_dist], axis=1) + ], + axis=0 + ) + original_dist = np.power(original_dist, 2).astype(np.float32) + original_dist = np.transpose( + 1. * original_dist / np.max(original_dist, axis=0) + ) + V = np.zeros_like(original_dist).astype(np.float32) + initial_rank = np.argsort(original_dist).astype(np.int32) + + query_num = q_g_dist.shape[0] + gallery_num = q_g_dist.shape[0] + q_g_dist.shape[1] + all_num = gallery_num + + for i in range(all_num): + # k-reciprocal neighbors + forward_k_neigh_index = initial_rank[i, :k1 + 1] + backward_k_neigh_index = initial_rank[forward_k_neigh_index, :k1 + 1] + fi = np.where(backward_k_neigh_index == i)[0] + k_reciprocal_index = forward_k_neigh_index[fi] + k_reciprocal_expansion_index = k_reciprocal_index + for j in range(len(k_reciprocal_index)): + candidate = k_reciprocal_index[j] + candidate_forward_k_neigh_index = initial_rank[ + candidate, :int(np.around(k1 / 2.)) + 1] + candidate_backward_k_neigh_index = initial_rank[ + candidate_forward_k_neigh_index, :int(np.around(k1 / 2.)) + 1] + fi_candidate = np.where( + candidate_backward_k_neigh_index == candidate + )[0] + candidate_k_reciprocal_index = candidate_forward_k_neigh_index[ + fi_candidate] + if len( + np. + intersect1d(candidate_k_reciprocal_index, k_reciprocal_index) + ) > 2. / 3 * len(candidate_k_reciprocal_index): + k_reciprocal_expansion_index = np.append( + k_reciprocal_expansion_index, candidate_k_reciprocal_index + ) + + k_reciprocal_expansion_index = np.unique(k_reciprocal_expansion_index) + weight = np.exp(-original_dist[i, k_reciprocal_expansion_index]) + V[i, k_reciprocal_expansion_index] = 1. * weight / np.sum(weight) + original_dist = original_dist[:query_num, ] + if k2 != 1: + V_qe = np.zeros_like(V, dtype=np.float32) + for i in range(all_num): + V_qe[i, :] = np.mean(V[initial_rank[i, :k2], :], axis=0) + V = V_qe + del V_qe + del initial_rank + invIndex = [] + for i in range(gallery_num): + invIndex.append(np.where(V[:, i] != 0)[0]) + + jaccard_dist = np.zeros_like(original_dist, dtype=np.float32) + + for i in range(query_num): + temp_min = np.zeros(shape=[1, gallery_num], dtype=np.float32) + indNonZero = np.where(V[i, :] != 0)[0] + indImages = [] + indImages = [invIndex[ind] for ind in indNonZero] + for j in range(len(indNonZero)): + temp_min[0, indImages[j]] = temp_min[0, indImages[j]] + np.minimum( + V[i, indNonZero[j]], V[indImages[j], indNonZero[j]] + ) + jaccard_dist[i] = 1 - temp_min / (2.-temp_min) + + final_dist = jaccard_dist * (1-lambda_value) + original_dist*lambda_value + del original_dist + del V + del jaccard_dist + final_dist = final_dist[:query_num, query_num:] + return final_dist diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py new file mode 100644 index 0000000000..d30752e112 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/tools.py @@ -0,0 +1,189 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import os +import sys +import json +import time +import errno +import numpy as np +import random +import os.path as osp +import warnings +import PIL +import torch +from PIL import Image + +__all__ = [ + 'mkdir_if_missing', 'check_isfile', 'read_json', 'write_json', + 'set_random_seed', 'download_url', 'read_image', 'collect_env_info', + 'listdir_nohidden' +] + + +def mkdir_if_missing(dirname): + """Creates dirname if it is missing.""" + if not osp.exists(dirname): + try: + os.makedirs(dirname) + except OSError as e: + if e.errno != errno.EEXIST: + raise + + +def check_isfile(fpath): + """Checks if the given path is a file. + + Args: + fpath (str): file path. + + Returns: + bool + """ + isfile = osp.isfile(fpath) + if not isfile: + warnings.warn('No file found at "{}"'.format(fpath)) + return isfile + + +def read_json(fpath): + """Reads json file from a path.""" + with open(fpath, 'r') as f: + obj = json.load(f) + return obj + + +def write_json(obj, fpath): + """Writes to a json file.""" + mkdir_if_missing(osp.dirname(fpath)) + with open(fpath, 'w') as f: + json.dump(obj, f, indent=4, separators=(',', ': ')) + + +def set_random_seed(seed): + random.seed(seed) + np.random.seed(seed) + torch.manual_seed(seed) + + +def download_url(url, dst): + """Downloads file from a url to a destination. + + Args: + url (str): url to download file. + dst (str): destination path. + """ + from six.moves import urllib + print('* url="{}"'.format(url)) + print('* destination="{}"'.format(dst)) + + def _reporthook(count, block_size, total_size): + global start_time + if count == 0: + start_time = time.time() + return + duration = time.time() - start_time + progress_size = int(count * block_size) + speed = int(progress_size / (1024*duration)) + percent = int(count * block_size * 100 / total_size) + sys.stdout.write( + '\r...%d%%, %d MB, %d KB/s, %d seconds passed' % + (percent, progress_size / (1024*1024), speed, duration) + ) + sys.stdout.flush() + + urllib.request.urlretrieve(url, dst, _reporthook) + sys.stdout.write('\n') + + +def read_image(path): + """Reads image from path using ``PIL.Image``. + + Args: + path (str): path to an image. + + Returns: + PIL image + """ + got_img = False + if not osp.exists(path): + raise IOError('"{}" does not exist'.format(path)) + while not got_img: + try: + img = Image.open(path).convert('RGB') + got_img = True + except IOError: + print( + 'IOError incurred when reading "{}". Will redo. Don\'t worry. Just chill.' + .format(path) + ) + return img + + +def collect_env_info(): + """Returns env info as a string. + + Code source: github.com/facebookresearch/maskrcnn-benchmark + """ + from torch.utils.collect_env import get_pretty_env_info + env_str = get_pretty_env_info() + env_str += '\n Pillow ({})'.format(PIL.__version__) + return env_str + + +def listdir_nohidden(path, sort=False): + """List non-hidden items in a directory. + + Args: + path (str): directory path. + sort (bool): sort the items. + """ + items = [f for f in os.listdir(path) if not f.startswith('.')] + if sort: + items.sort() + return items diff --git a/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py new file mode 100644 index 0000000000..f9d5d6f12d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/torchreid/utils/torchtools.py @@ -0,0 +1,363 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from __future__ import division, print_function, absolute_import +import pickle +import shutil +import os.path as osp +import warnings +from functools import partial +from collections import OrderedDict +import torch +import torch.nn as nn + +from .tools import mkdir_if_missing + +__all__ = [ + 'save_checkpoint', 'load_checkpoint', 'resume_from_checkpoint', + 'open_all_layers', 'open_specified_layers', 'count_num_param', + 'load_pretrained_weights' +] + + +def save_checkpoint( + state, save_dir, is_best=False, remove_module_from_keys=False +): + r"""Saves checkpoint. + + Args: + state (dict): dictionary. + save_dir (str): directory to save checkpoint. + is_best (bool, optional): if True, this checkpoint will be copied and named + ``model-best.pth.tar``. Default is False. + remove_module_from_keys (bool, optional): whether to remove "module." + from layer names. Default is False. + + Examples:: + >>> state = { + >>> 'state_dict': model.state_dict(), + >>> 'epoch': 10, + >>> 'rank1': 0.5, + >>> 'optimizer': optimizer.state_dict() + >>> } + >>> save_checkpoint(state, 'log/my_model') + """ + mkdir_if_missing(save_dir) + if remove_module_from_keys: + # remove 'module.' in state_dict's keys + state_dict = state['state_dict'] + new_state_dict = OrderedDict() + for k, v in state_dict.items(): + if k.startswith('module.'): + k = k[7:] + new_state_dict[k] = v + state['state_dict'] = new_state_dict + # save + epoch = state['epoch'] + fpath = osp.join(save_dir, 'model.pth.tar-' + str(epoch)) + torch.save(state, fpath) + print('Checkpoint saved to "{}"'.format(fpath)) + if is_best: + shutil.copy(fpath, osp.join(osp.dirname(fpath), 'model-best.pth.tar')) + + +def load_checkpoint(fpath): + r"""Loads checkpoint. + + ``UnicodeDecodeError`` can be well handled, which means + python2-saved files can be read from python3. + + Args: + fpath (str): path to checkpoint. + + Returns: + dict + + Examples:: + >>> from torchreid.utils import load_checkpoint + >>> fpath = 'log/my_model/model.pth.tar-10' + >>> checkpoint = load_checkpoint(fpath) + """ + if fpath is None: + raise ValueError('File path is None') + if not osp.exists(fpath): + raise FileNotFoundError('File is not found at "{}"'.format(fpath)) + map_location = None if torch.cuda.is_available() else 'cpu' + try: + checkpoint = torch.load(fpath, map_location=map_location) + except UnicodeDecodeError: + pickle.load = partial(pickle.load, encoding="latin1") + pickle.Unpickler = partial(pickle.Unpickler, encoding="latin1") + checkpoint = torch.load( + fpath, pickle_module=pickle, map_location=map_location + ) + except Exception: + print('Unable to load checkpoint from "{}"'.format(fpath)) + raise + return checkpoint + + +def resume_from_checkpoint(fpath, model, optimizer=None, scheduler=None): + r"""Resumes training from a checkpoint. + + This will load (1) model weights and (2) ``state_dict`` + of optimizer if ``optimizer`` is not None. + + Args: + fpath (str): path to checkpoint. + model (nn.Module): model. + optimizer (Optimizer, optional): an Optimizer. + scheduler (LRScheduler, optional): an LRScheduler. + + Returns: + int: start_epoch. + + Examples:: + >>> from torchreid.utils import resume_from_checkpoint + >>> fpath = 'log/my_model/model.pth.tar-10' + >>> start_epoch = resume_from_checkpoint( + >>> fpath, model, optimizer, scheduler + >>> ) + """ + print('Loading checkpoint from "{}"'.format(fpath)) + checkpoint = load_checkpoint(fpath) + model.load_state_dict(checkpoint['state_dict']) + print('Loaded model weights') + if optimizer is not None and 'optimizer' in checkpoint.keys(): + optimizer.load_state_dict(checkpoint['optimizer']) + print('Loaded optimizer') + if scheduler is not None and 'scheduler' in checkpoint.keys(): + scheduler.load_state_dict(checkpoint['scheduler']) + print('Loaded scheduler') + start_epoch = checkpoint['epoch'] + print('Last epoch = {}'.format(start_epoch)) + if 'rank1' in checkpoint.keys(): + print('Last rank1 = {:.1%}'.format(checkpoint['rank1'])) + return start_epoch + + +def adjust_learning_rate( + optimizer, + base_lr, + epoch, + stepsize=20, + gamma=0.1, + linear_decay=False, + final_lr=0, + max_epoch=100 +): + r"""Adjusts learning rate. + + Deprecated. + """ + if linear_decay: + # linearly decay learning rate from base_lr to final_lr + frac_done = epoch / max_epoch + lr = frac_done*final_lr + (1.-frac_done) * base_lr + else: + # decay learning rate by gamma for every stepsize + lr = base_lr * (gamma**(epoch // stepsize)) + + for param_group in optimizer.param_groups: + param_group['lr'] = lr + + +def set_bn_to_eval(m): + r"""Sets BatchNorm layers to eval mode.""" + # 1. no update for running mean and var + # 2. scale and shift parameters are still trainable + classname = m.__class__.__name__ + if classname.find('BatchNorm') != -1: + m.eval() + + +def open_all_layers(model): + r"""Opens all layers in model for training. + + Examples:: + >>> from torchreid.utils import open_all_layers + >>> open_all_layers(model) + """ + model.train() + for p in model.parameters(): + p.requires_grad = True + + +def open_specified_layers(model, open_layers): + r"""Opens specified layers in model for training while keeping + other layers frozen. + + Args: + model (nn.Module): neural net model. + open_layers (str or list): layers open for training. + + Examples:: + >>> from torchreid.utils import open_specified_layers + >>> # Only model.classifier will be updated. + >>> open_layers = 'classifier' + >>> open_specified_layers(model, open_layers) + >>> # Only model.fc and model.classifier will be updated. + >>> open_layers = ['fc', 'classifier'] + >>> open_specified_layers(model, open_layers) + """ + if isinstance(model, nn.DataParallel): + model = model.module + elif isinstance(model, nn.parallel.DistributedDataParallel): + model = model.module + + if isinstance(open_layers, str): + open_layers = [open_layers] + + for layer in open_layers: + assert hasattr( + model, layer + ), '"{}" is not an attribute of the model, please provide the correct name'.format( + layer + ) + + for name, module in model.named_children(): + if name in open_layers: + module.train() + for p in module.parameters(): + p.requires_grad = True + else: + module.eval() + for p in module.parameters(): + p.requires_grad = False + + +def count_num_param(model): + r"""Counts number of parameters in a model while ignoring ``self.classifier``. + + Args: + model (nn.Module): network model. + + Examples:: + >>> from torchreid.utils import count_num_param + >>> model_size = count_num_param(model) + + .. warning:: + + This method is deprecated in favor of + ``torchreid.utils.compute_model_complexity``. + """ + warnings.warn( + 'This method is deprecated and will be removed in the future.' + ) + + num_param = sum(p.numel() for p in model.parameters()) + + if isinstance(model, nn.DataParallel): + model = model.module + + if hasattr(model, + 'classifier') and isinstance(model.classifier, nn.Module): + # we ignore the classifier because it is unused at test time + num_param -= sum(p.numel() for p in model.classifier.parameters()) + + return num_param + + +def load_pretrained_weights(model, weight_path, ignore_classifier=False): + r"""Loads pretrianed weights to model. + + Features:: + - Incompatible layers (unmatched in name or size) will be ignored. + - Can automatically deal with keys containing "module.". + + Args: + model (nn.Module): network model. + weight_path (str): path to pretrained weights. + + Examples:: + >>> from torchreid.utils import load_pretrained_weights + >>> weight_path = 'log/my_model/model-best.pth.tar' + >>> load_pretrained_weights(model, weight_path) + """ + checkpoint = load_checkpoint(weight_path) + if 'state_dict' in checkpoint: + state_dict = checkpoint['state_dict'] + else: + state_dict = checkpoint + + model_dict = model.state_dict() + new_state_dict = OrderedDict() + matched_layers, discarded_layers = [], [] + + for k, v in state_dict.items(): + if k.startswith('module.'): + k = k[7:] # discard module. + + if ignore_classifier and k.startswith('classifier'): + continue + + if k in model_dict and model_dict[k].size() == v.size(): + new_state_dict[k] = v + matched_layers.append(k) + else: + discarded_layers.append(k) + + model_dict.update(new_state_dict) + model.load_state_dict(model_dict) + + if len(matched_layers) == 0: + warnings.warn( + 'The pretrained weights "{}" cannot be loaded, ' + 'please check the key names manually ' + '(** ignored and continue **)'.format(weight_path) + ) + else: + print( + 'Successfully loaded pretrained weights from "{}"'. + format(weight_path) + ) + if len(discarded_layers) > 0: + print( + '** The following layers are discarded ' + 'due to unmatched keys or layer size: {}'. + format(discarded_layers) + ) -- Gitee From d10e2ab5e88d31bd58b34b3338dc885fba5bc6b6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:10 +0000 Subject: [PATCH 14/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/.flake8?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/.flake8 | 18 ------------------ 1 file changed, 18 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/.flake8 diff --git a/PyTorch/contrib/cv/classification/OSNet/.flake8 b/PyTorch/contrib/cv/classification/OSNet/.flake8 deleted file mode 100644 index 4fc103cb10..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/.flake8 +++ /dev/null @@ -1,18 +0,0 @@ -[flake8] -ignore = - # At least two spaces before inline comment - E261, - # Line lengths are recommended to be no greater than 79 characters - E501, - # Missing whitespace around arithmetic operator - E226, - # Blank line contains whitespace - W293, - # Do not use bare 'except' - E722, - # Line break after binary operator - W504, - # isort found an import in the wrong position - I001 -max-line-length = 79 -exclude = __init__.py, build, torchreid/metrics/rank_cylib/ \ No newline at end of file -- Gitee From 6b452513af36475ce3a3f08d9306ac0f0494607a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:33 +0000 Subject: [PATCH 15/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/.gitignore?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/.gitignore | 146 ------------------ 1 file changed, 146 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/.gitignore diff --git a/PyTorch/contrib/cv/classification/OSNet/.gitignore b/PyTorch/contrib/cv/classification/OSNet/.gitignore deleted file mode 100644 index 7f6c2b6264..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/.gitignore +++ /dev/null @@ -1,146 +0,0 @@ -# Byte-compiled / optimized / DLL files -__pycache__/ -*.py[cod] -*$py.class - -nohup.out -inference/ -fusion_result.json -kernel_meta/ -cann_profiling/ -*.onnx -# C extensions -*.so - -# Distribution / packaging -.Python -build/ -develop-eggs/ -dist/ -downloads/ -eggs/ -.eggs/ -lib/ -lib64/ -parts/ -sdist/ -var/ -wheels/ -pip-wheel-metadata/ -share/python-wheels/ -*.egg-info/ -.installed.cfg -*.egg -MANIFEST - -# PyInstaller -# Usually these files are written by a python script from a template -# before PyInstaller builds the exe, so as to inject date/other infos into it. -*.manifest -*.spec - -# Installer logs -pip-log.txt -pip-delete-this-directory.txt - -# Unit test / coverage reports -htmlcov/ -.tox/ -.nox/ -.coverage -.coverage.* -.cache -nosetests.xml -coverage.xml -*.cover -.hypothesis/ -.pytest_cache/ - -# Translations -*.mo -*.pot - -# Django stuff: -*.log -local_settings.py -db.sqlite3 - -# Flask stuff: -instance/ -.webassets-cache - -# Scrapy stuff: -.scrapy - -# Sphinx documentation -docs/_build/ - -# PyBuilder -target/ - -# Jupyter Notebook -.ipynb_checkpoints - -# IPython -profile_default/ -ipython_config.py - -# pyenv -.python-version - -# pipenv -# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control. -# However, in case of collaboration, if having platform-specific dependencies or dependencies -# having no cross-platform support, pipenv may install dependencies that don't work, or not -# install all needed dependencies. -#Pipfile.lock - -# celery beat schedule file -celerybeat-schedule - -# SageMath parsed files -*.sage.py - -# Environments -.env -.venv -env/ -venv/ -ENV/ -env.bak/ -venv.bak/ - -# Spyder project settings -.spyderproject -.spyproject - -# Rope project settings -.ropeproject - -# mkdocs documentation -/site - -# mypy -.mypy_cache/ -.dmypy.json -dmypy.json - -# Pyre type checker -.pyre/ - -# Cython eval code -*.c -*.html - -# OS X -.DS_Store -.Spotlight-V100 -.Trashes -._* - -# ReID -reid-data/ -log/ -saved-models/ -model-zoo/ -debug* -- Gitee From 221fc2ce21c7530a865efe2ce30449f2b96aa90f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:40 +0000 Subject: [PATCH 16/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/.isort.cfg?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- PyTorch/contrib/cv/classification/OSNet/.isort.cfg | 10 ---------- 1 file changed, 10 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/.isort.cfg diff --git a/PyTorch/contrib/cv/classification/OSNet/.isort.cfg b/PyTorch/contrib/cv/classification/OSNet/.isort.cfg deleted file mode 100644 index 8039326b5c..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/.isort.cfg +++ /dev/null @@ -1,10 +0,0 @@ -[isort] -line_length=79 -multi_line_output=3 -length_sort=true -known_standard_library=numpy,setuptools -known_myself=torchreid -known_third_party=matplotlib,cv2,torch,torchvision,PIL,yacs -no_lines_before=STDLIB,THIRDPARTY -sections=FUTURE,STDLIB,THIRDPARTY,myself,FIRSTPARTY,LOCALFOLDER -default_section=FIRSTPARTY \ No newline at end of file -- Gitee From 96887b5e981de4761ad13cad335b0947ea52903a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:46 +0000 Subject: [PATCH 17/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/.style.yapf?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- PyTorch/contrib/cv/classification/OSNet/.style.yapf | 7 ------- 1 file changed, 7 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/.style.yapf diff --git a/PyTorch/contrib/cv/classification/OSNet/.style.yapf b/PyTorch/contrib/cv/classification/OSNet/.style.yapf deleted file mode 100644 index 29d8e52cc8..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/.style.yapf +++ /dev/null @@ -1,7 +0,0 @@ -[style] -BASED_ON_STYLE = pep8 -BLANK_LINE_BEFORE_NESTED_CLASS_OR_DEF = true -SPLIT_BEFORE_EXPRESSION_AFTER_OPENING_PAREN = true -DEDENT_CLOSING_BRACKETS = true -SPACES_BEFORE_COMMENT = 1 -ARITHMETIC_PRECEDENCE_INDICATION = true \ No newline at end of file -- Gitee From 5357d23e6e0ad515b36864e73e2164dd5f381ca6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:52 +0000 Subject: [PATCH 18/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/LICENSE?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/LICENSE | 21 ------------------- 1 file changed, 21 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/LICENSE diff --git a/PyTorch/contrib/cv/classification/OSNet/LICENSE b/PyTorch/contrib/cv/classification/OSNet/LICENSE deleted file mode 100644 index d2bcb88271..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/LICENSE +++ /dev/null @@ -1,21 +0,0 @@ -MIT License - -Copyright (c) 2018 Kaiyang Zhou - -Permission is hereby granted, free of charge, to any person obtaining a copy -of this software and associated documentation files (the "Software"), to deal -in the Software without restriction, including without limitation the rights -to use, copy, modify, merge, publish, distribute, sublicense, and/or sell -copies of the Software, and to permit persons to whom the Software is -furnished to do so, subject to the following conditions: - -The above copyright notice and this permission notice shall be included in all -copies or substantial portions of the Software. - -THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE -AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER -LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, -OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE -SOFTWARE. -- Gitee From ebdd672ef4609669b2f06645f831fe54e8b680f9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:30:58 +0000 Subject: [PATCH 19/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/README.md?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/README.md | 72 ------------------- 1 file changed, 72 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/README.md diff --git a/PyTorch/contrib/cv/classification/OSNet/README.md b/PyTorch/contrib/cv/classification/OSNet/README.md deleted file mode 100644 index 2528236875..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/README.md +++ /dev/null @@ -1,72 +0,0 @@ -# OSNet - -This implements training of OSNet on the Market-1501 dataset, mainly modified from [KaiyangZhou/deep-person-reid](https://github.com/KaiyangZhou/deep-person-reid). - -## OSNet Detail - -As of the current date, Ascend-Pytorch is still inefficient for contiguous operations.Therefore, OSNet is re-implemented using semantics such as custom OP. - - -## Requirements - -- Install PyTorch ([pytorch.org](http://pytorch.org)) - - -- `pip install -r requirements.txt` - -- Install torchreid - - - ~~~python - python setup.py develop - ~~~ - -- Download the Market-1501 dataset from https://paperswithcode.com/dataset/market-1501 - - - ~~~shell - unzip Market-1501-v15.09.15.zip - ~~~ - -- Move Market-1501 dataset to 'reid-data' path - - - ~~~shell - mkdir path_to_osnet/reid-data/ - mv Market-1501-v15.09.15 path_to_osnet/reid-data/market1501 - ~~~ -## Training - -To train a model, run `main.py` with the desired model architecture and the path to the ImageNet dataset: - -```bash -# training 1p accuracy -bash test/train_full_1p.sh - -# training 1p performance -bash test/train_performance_1p.sh - -# training 8p accuracy -bash test/train_full_8p.sh - -# training 8p performance -bash test/train_performance_8p.sh - -# finetuning -bash test/train_finetune_1p.sh --data_path=real_data_path --weight=real_weight_path - -# Online inference demo -python demo.py -## 备注： 识别前后图片保存到 `inference/` 文件夹下 - -# To ONNX -python pthtar2onnx.py -``` - -## OSNet training result - - -| | mAP | AMP_Type | Epochs | FPS | -| :----: | :--: | :------: | :----: | :------: | -| 1p-GPU | - | O2 | 1 | 371.383 | -| 1p-NPU | - | O2 | 1 | 366.464 | -| 8p-GPU | 80.3 | O2 | 350 | 1045.535 | -| 8p-NPU | 80.2 | O2 | 350 | 1091.358 | - -- Gitee From 11a58c98ab2d11249ccfe8ed54a1c746d179e07e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:04 +0000 Subject: [PATCH 20/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/default=5Fconfig.py?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/default_config.py | 272 ------------------ 1 file changed, 272 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/default_config.py diff --git a/PyTorch/contrib/cv/classification/OSNet/default_config.py b/PyTorch/contrib/cv/classification/OSNet/default_config.py deleted file mode 100644 index a2e22fe44d..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/default_config.py +++ /dev/null @@ -1,272 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -from yacs.config import CfgNode as CN - - -def get_default_config(): - cfg = CN() - - # model - cfg.model = CN() - cfg.model.name = 'resnet50' - cfg.model.pretrained = True # automatically load pretrained model weights if available - cfg.model.load_weights = '' # path to model weights - cfg.model.resume = '' # path to checkpoint for resume training - - # data - cfg.data = CN() - cfg.data.type = 'image' - cfg.data.root = 'reid-data' - cfg.data.sources = ['market1501'] - cfg.data.targets = ['market1501'] - cfg.data.workers = 4 # number of data loading workers - cfg.data.split_id = 0 # split index - cfg.data.height = 256 # image height - cfg.data.width = 128 # image width - cfg.data.combineall = False # combine train, query and gallery for training - cfg.data.transforms = ['random_flip'] # data augmentation - cfg.data.k_tfm = 1 # number of times to apply augmentation to an image independently - cfg.data.norm_mean = [0.485, 0.456, 0.406] # default is imagenet mean - cfg.data.norm_std = [0.229, 0.224, 0.225] # default is imagenet std - cfg.data.save_dir = 'log' # path to save log - cfg.data.load_train_targets = False # load training set from target dataset - - # specific datasets - cfg.market1501 = CN() - cfg.market1501.use_500k_distractors = False # add 500k distractors to the gallery set for market1501 - cfg.cuhk03 = CN() - cfg.cuhk03.labeled_images = False # use labeled images, if False, use detected images - cfg.cuhk03.classic_split = False # use classic split by Li et al. CVPR14 - cfg.cuhk03.use_metric_cuhk03 = False # use cuhk03's metric for evaluation - - # sampler - cfg.sampler = CN() - cfg.sampler.train_sampler = 'RandomSampler' # sampler for source train loader - cfg.sampler.train_sampler_t = 'RandomSampler' # sampler for target train loader - cfg.sampler.num_instances = 4 # number of instances per identity for RandomIdentitySampler - cfg.sampler.num_cams = 1 # number of cameras to sample in a batch (for RandomDomainSampler) - cfg.sampler.num_datasets = 1 # number of datasets to sample in a batch (for RandomDatasetSampler) - - # video reid setting - cfg.video = CN() - cfg.video.seq_len = 15 # number of images to sample in a tracklet - cfg.video.sample_method = 'evenly' # how to sample images from a tracklet - cfg.video.pooling_method = 'avg' # how to pool features over a tracklet - - # train - cfg.train = CN() - cfg.train.optim = 'adam' - cfg.train.lr = 0.0003 - cfg.train.weight_decay = 5e-4 - cfg.train.max_epoch = 60 - cfg.train.start_epoch = 0 - cfg.train.batch_size = 32 - cfg.train.fixbase_epoch = 0 # number of epochs to fix base layers - cfg.train.open_layers = [ - 'classifier' - ] # layers for training while keeping others frozen - cfg.train.staged_lr = False # set different lr to different layers - cfg.train.new_layers = ['classifier'] # newly added layers with default lr - cfg.train.base_lr_mult = 0.1 # learning rate multiplier for base layers - cfg.train.lr_scheduler = 'single_step' - cfg.train.stepsize = [20] # stepsize to decay learning rate - cfg.train.gamma = 0.1 # learning rate decay multiplier - cfg.train.print_freq = 20 # print frequency - cfg.train.seed = 1 # random seed - - # optimizer - cfg.sgd = CN() - cfg.sgd.momentum = 0.9 # momentum factor for sgd and rmsprop - cfg.sgd.dampening = 0. # dampening for momentum - cfg.sgd.nesterov = False # Nesterov momentum - cfg.rmsprop = CN() - cfg.rmsprop.alpha = 0.99 # smoothing constant - cfg.adam = CN() - cfg.adam.beta1 = 0.9 # exponential decay rate for first moment - cfg.adam.beta2 = 0.999 # exponential decay rate for second moment - - # loss - cfg.loss = CN() - cfg.loss.name = 'softmax' - cfg.loss.softmax = CN() - cfg.loss.softmax.label_smooth = True # use label smoothing regularizer - cfg.loss.triplet = CN() - cfg.loss.triplet.margin = 0.3 # distance margin - cfg.loss.triplet.weight_t = 1. # weight to balance hard triplet loss - cfg.loss.triplet.weight_x = 0. # weight to balance cross entropy loss - - # test - cfg.test = CN() - cfg.test.batch_size = 100 - cfg.test.dist_metric = 'euclidean' # distance metric, ['euclidean', 'cosine'] - cfg.test.normalize_feature = False # normalize feature vectors before computing distance - cfg.test.ranks = [1, 5, 10, 20] # cmc ranks - cfg.test.evaluate = False # test only - cfg.test.eval_freq = -1 # evaluation frequency (-1 means to only test after training) - cfg.test.start_eval = 0 # start to evaluate after a specific epoch - cfg.test.rerank = False # use person re-ranking - cfg.test.visrank = False # visualize ranked results (only available when cfg.test.evaluate=True) - cfg.test.visrank_topk = 10 # top-k ranks to visualize - - return cfg - - -def imagedata_kwargs(cfg): - return { - 'root': cfg.data.root, - 'sources': cfg.data.sources, - 'targets': cfg.data.targets, - 'height': cfg.data.height, - 'width': cfg.data.width, - 'transforms': cfg.data.transforms, - 'k_tfm': cfg.data.k_tfm, - 'norm_mean': cfg.data.norm_mean, - 'norm_std': cfg.data.norm_std, - 'use_gpu': cfg.use_gpu, - 'split_id': cfg.data.split_id, - 'combineall': cfg.data.combineall, - 'load_train_targets': cfg.data.load_train_targets, - 'batch_size_train': cfg.train.batch_size, - 'batch_size_test': cfg.test.batch_size, - 'workers': cfg.data.workers, - 'num_instances': cfg.sampler.num_instances, - 'num_cams': cfg.sampler.num_cams, - 'num_datasets': cfg.sampler.num_datasets, - 'train_sampler': cfg.sampler.train_sampler, - 'train_sampler_t': cfg.sampler.train_sampler_t, - # image dataset specific - 'cuhk03_labeled': cfg.cuhk03.labeled_images, - 'cuhk03_classic_split': cfg.cuhk03.classic_split, - 'market1501_500k': cfg.market1501.use_500k_distractors, - 'device_num': cfg.device_num, - } - - -def videodata_kwargs(cfg): - return { - 'root': cfg.data.root, - 'sources': cfg.data.sources, - 'targets': cfg.data.targets, - 'height': cfg.data.height, - 'width': cfg.data.width, - 'transforms': cfg.data.transforms, - 'norm_mean': cfg.data.norm_mean, - 'norm_std': cfg.data.norm_std, - 'use_gpu': cfg.use_gpu, - 'split_id': cfg.data.split_id, - 'combineall': cfg.data.combineall, - 'batch_size_train': cfg.train.batch_size, - 'batch_size_test': cfg.test.batch_size, - 'workers': cfg.data.workers, - 'num_instances': cfg.sampler.num_instances, - 'num_cams': cfg.sampler.num_cams, - 'num_datasets': cfg.sampler.num_datasets, - 'train_sampler': cfg.sampler.train_sampler, - # video dataset specific - 'seq_len': cfg.video.seq_len, - 'sample_method': cfg.video.sample_method - } - - -def optimizer_kwargs(cfg): - return { - 'optim': cfg.train.optim, - 'lr': cfg.train.lr, - 'weight_decay': cfg.train.weight_decay, - 'momentum': cfg.sgd.momentum, - 'sgd_dampening': cfg.sgd.dampening, - 'sgd_nesterov': cfg.sgd.nesterov, - 'rmsprop_alpha': cfg.rmsprop.alpha, - 'adam_beta1': cfg.adam.beta1, - 'adam_beta2': cfg.adam.beta2, - 'staged_lr': cfg.train.staged_lr, - 'new_layers': cfg.train.new_layers, - 'base_lr_mult': cfg.train.base_lr_mult - } - - -def lr_scheduler_kwargs(cfg): - return { - 'lr_scheduler': cfg.train.lr_scheduler, - 'stepsize': cfg.train.stepsize, - 'gamma': cfg.train.gamma, - 'max_epoch': cfg.train.max_epoch - } - - -def engine_run_kwargs(cfg): - return { - 'save_dir': cfg.data.save_dir, - 'max_epoch': cfg.train.max_epoch, - 'start_epoch': cfg.train.start_epoch, - 'fixbase_epoch': cfg.train.fixbase_epoch, - 'open_layers': cfg.train.open_layers, - 'start_eval': cfg.test.start_eval, - 'eval_freq': cfg.test.eval_freq, - 'test_only': cfg.test.evaluate, - 'print_freq': cfg.train.print_freq, - 'dist_metric': cfg.test.dist_metric, - 'normalize_feature': cfg.test.normalize_feature, - 'visrank': cfg.test.visrank, - 'visrank_topk': cfg.test.visrank_topk, - 'use_metric_cuhk03': cfg.cuhk03.use_metric_cuhk03, - 'ranks': cfg.test.ranks, - 'rerank': cfg.test.rerank - } - -def evaluate_kwargs(cfg): - return { - 'dist_metric': cfg.test.dist_metric, - 'normalize_feature': cfg.test.normalize_feature, - 'visrank': cfg.test.visrank, - 'visrank_topk': cfg.test.visrank_topk, - 'save_dir': cfg.data.save_dir, - 'use_metric_cuhk03': cfg.cuhk03.use_metric_cuhk03, - 'ranks': cfg.test.ranks, - 'rerank': cfg.test.rerank - } \ No newline at end of file -- Gitee From 48bb0b3354c089f2124b4fb7f0974a934fae31e0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:09 +0000 Subject: [PATCH 21/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/demo.py?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/demo.py | 187 ------------------ 1 file changed, 187 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/demo.py diff --git a/PyTorch/contrib/cv/classification/OSNet/demo.py b/PyTorch/contrib/cv/classification/OSNet/demo.py deleted file mode 100644 index deb7c533fe..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/demo.py +++ /dev/null @@ -1,187 +0,0 @@ -#!/usr/bin/env python3 -# Copyright 2020 Huawei Technologies Co., Ltd -# -# Licensed under the BSD 3-Clause License (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# https://spdx.org/licenses/BSD-3-Clause.html -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. - -import torch -import numpy as np -import os.path as osp -import os -import argparse -import torchreid -from torchreid.data.datasets.image.market1501 import Market1501 -from torchreid.utils import load_pretrained_weights -from torchreid import metrics - -if not os.path.exists("inference"): - os.makedirs("inference") -os.system('rm -f inference/*') - -parser = argparse.ArgumentParser( - formatter_class=argparse.ArgumentDefaultsHelpFormatter -) -# data -parser.add_argument('-d', '--data_path', type=str, default='./reid-data/') -parser.add_argument('--device', type=str, default='cpu') -parser.add_argument('--checkpoint', type=str, default='log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350') -parser.add_argument('--config_file', type=str, default='configs/osnet_x1_0_trained_from_scratch.yaml') - -args = parser.parse_args() - -os.environ['device'] = args.device - -def build_model(): - # Create model - model = torchreid.models.build_model( - name="osnet_x1_0", - num_classes=751, - loss="softmax", - pretrained=False, - use_gpu=False - ) - load_pretrained_weights(model, args.checkpoint) - if os.environ['device'] == 'npu': - model = model.to("npu:0") - elif os.environ['device'] == 'gpu': - model = model.to("cuda:0") - model.eval() - return model - - -def get_raw_data(): - name, root = "market1501", args.data_path - - param = { - "root": root, - 'sources': ['market1501'], - 'targets': ['market1501'], - 'height': 256, - 'width': 128, - 'transforms': ['random_flip', 'random_crop', 'random_patch'], - 'k_tfm': 1, - 'norm_mean': [0.485, 0.456, 0.406], - 'norm_std': [0.229, 0.224, 0.225], - 'use_gpu': False, - 'split_id': 0, - 'combineall': False, - 'load_train_targets': False, - 'batch_size_train': 32, - 'batch_size_test': 32, - 'workers': 4, - 'num_instances': 4, - 'num_cams': 1, - 'num_datasets': 1, - 'train_sampler': "RandomSampler", - 'train_sampler_t': "RandomSampler", - # image dataset specific - 'cuhk03_labeled': False, - 'cuhk03_classic_split': False, - 'market1501_500k': False, - } - - datamanager = torchreid.data.ImageDataManager(**param) - - query_loader = datamanager.test_loader['market1501']['query'] - gallery_loader = datamanager.test_loader['market1501']['gallery'] - - data = next(iter(query_loader)) - - fnames = data['impath'] - pids = data['pid'] - imgs = data['img'] - camids = data['camid'] - - img = imgs[24] - pid = pids[24] - camid = camids[24] - fname = fnames[24] - img = torch.unsqueeze(img, dim=0) - - return datamanager, gallery_loader, img, fname, pid - -def parse_data_for_eval(data): - fnames = data['impath'] - imgs = data['img'] - pids = data['pid'] - camids = data['camid'] - return fnames, imgs, pids, camids - -def feature_extraction(model, imgs): - if os.environ['device'] == 'gpu': - imgs = imgs.cuda() - elif os.environ['device'] == 'npu': - imgs = imgs.npu() - features = model(imgs) - features = features.cpu().clone() - return features - -def find_imgs_with_id(id, data_loader): - f_, pids_, camids_, f_names_ = [], [], [], [] - for batch_idx, data in enumerate(data_loader): - fnames, imgs, pids, camids = parse_data_for_eval(data) - for fname, img, pid, camid in zip(fnames, imgs, pids, camids): - if pid == id: - img = torch.unsqueeze(img, dim=0) - f_.append(img) - pids_.append(pid) - camids_.append(camid) - f_names_.append(fname) - f_ = torch.cat(f_, dim=0) - return f_, pids_, camids_, f_names_ - -def feature_extraction_single(model, imgs): - if os.environ['device'] == 'gpu': - imgs = imgs.cuda() - elif os.environ['device'] == 'npu': - imgs = imgs.npu() - features = model(imgs) - features = features.cpu().clone() - return features - -def save_image(fname): - img_name = osp.basename(fname) - command = "cp %s ./inference/%s" % (fname, img_name) - os.system(command) - - -if __name__ == '__main__': - data_path = args.data_path - print("load dataset") - datamanager, gallery_loader, img, fname, pid = get_raw_data() - print("find a img in gallery with id %d" % pid) - imgs_gallery, pids, camids, f_names = find_imgs_with_id(pid.item(), gallery_loader) - - print("build model") - model = build_model() - print("extract img feature...") - img_feature = feature_extraction_single(model, img) - print("extract gallery feature...") - # gallery_feature = feature_extraction_single(model, img_gallery) - gallery_feature = feature_extraction(model, imgs_gallery) - - dist_metric = "euclidean" - print( - 'Computing distance matrix with metric={} ...'.format(dist_metric) - ) - distmat = metrics.compute_distance_matrix(img_feature, gallery_feature, dist_metric) - distmat = distmat.cpu().detach().numpy() - m, n = distmat.shape - indices = np.argsort(distmat, axis=1) - index = indices[0][0] - fname_gallery = f_names[index] - - save_image(fname) - save_image(fname_gallery) - print("query img saved to ./inference/%s" % osp.basename(fname)) - print("gallery img saved to ./inference/%s" % osp.basename(fname_gallery)) - \ No newline at end of file -- Gitee From 59e87efb7405b77a53f2e31b0eb093e7d79467b0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:16 +0000 Subject: [PATCH 22/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/main.py?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/main.py | 310 ------------------ 1 file changed, 310 deletions(-) delete mode 100755 PyTorch/contrib/cv/classification/OSNet/main.py diff --git a/PyTorch/contrib/cv/classification/OSNet/main.py b/PyTorch/contrib/cv/classification/OSNet/main.py deleted file mode 100755 index 6da72b3c69..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/main.py +++ /dev/null @@ -1,310 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -import sys -import time -import os.path as osp -import argparse -import torch -import torch.nn as nn -import os - -import torchreid -from torchreid.utils import ( - Logger, check_isfile, set_random_seed, collect_env_info, - resume_from_checkpoint, load_pretrained_weights, compute_model_complexity -) - -from default_config import ( - imagedata_kwargs, optimizer_kwargs, videodata_kwargs, engine_run_kwargs, - get_default_config, lr_scheduler_kwargs, evaluate_kwargs -) - -from apex import amp - - -def build_datamanager(cfg): - if cfg.data.type == 'image': - return torchreid.data.ImageDataManager(**imagedata_kwargs(cfg)) - else: - return torchreid.data.VideoDataManager(**videodata_kwargs(cfg)) - - -def build_engine(cfg, datamanager, model, optimizer, scheduler): - if cfg.data.type == 'image': - if cfg.loss.name == 'softmax': - engine = torchreid.engine.ImageSoftmaxEngine( - datamanager, - model, - optimizer=optimizer, - scheduler=scheduler, - use_gpu=cfg.use_gpu, - use_npu=cfg.use_npu, - label_smooth=cfg.loss.softmax.label_smooth, - use_amp=cfg.amp - ) - - else: - engine = torchreid.engine.ImageTripletEngine( - datamanager, - model, - optimizer=optimizer, - margin=cfg.loss.triplet.margin, - weight_t=cfg.loss.triplet.weight_t, - weight_x=cfg.loss.triplet.weight_x, - scheduler=scheduler, - use_gpu=cfg.use_gpu, - label_smooth=cfg.loss.softmax.label_smooth - ) - - else: - if cfg.loss.name == 'softmax': - engine = torchreid.engine.VideoSoftmaxEngine( - datamanager, - model, - optimizer=optimizer, - scheduler=scheduler, - use_gpu=cfg.use_gpu, - label_smooth=cfg.loss.softmax.label_smooth, - pooling_method=cfg.video.pooling_method - ) - - else: - engine = torchreid.engine.VideoTripletEngine( - datamanager, - model, - optimizer=optimizer, - margin=cfg.loss.triplet.margin, - weight_t=cfg.loss.triplet.weight_t, - weight_x=cfg.loss.triplet.weight_x, - scheduler=scheduler, - use_gpu=cfg.use_gpu, - label_smooth=cfg.loss.softmax.label_smooth - ) - - return engine - - -def reset_config(cfg, args): - if args.root: - cfg.data.root = args.root - if args.sources: - cfg.data.sources = args.sources - if args.targets: - cfg.data.targets = args.targets - if args.transforms: - cfg.data.transforms = args.transforms - - -def check_cfg(cfg): - if cfg.loss.name == 'triplet' and cfg.loss.triplet.weight_x == 0: - assert cfg.train.fixbase_epoch == 0, \ - 'The output of classifier is not included in the computational graph' - - -def main(): - parser = argparse.ArgumentParser( - formatter_class=argparse.ArgumentDefaultsHelpFormatter - ) - parser.add_argument( - '--config-file', type=str, default='', help='path to config file' - ) - parser.add_argument( - '-s', - '--sources', - type=str, - nargs='+', - help='source datasets (delimited by space)' - ) - parser.add_argument( - '-t', - '--targets', - type=str, - nargs='+', - help='target datasets (delimited by space)' - ) - parser.add_argument( - '--transforms', type=str, nargs='+', help='data augmentation' - ) - parser.add_argument( - '--root', type=str, default='', help='path to data root' - ) - parser.add_argument( - 'opts', - default=None, - nargs=argparse.REMAINDER, - help='Modify config options using the command-line' - ) - - parser.add_argument("--local_rank", type=int, default=0) - parser.add_argument('--device_num', default=-1, type=int, - help='device_num') - parser.add_argument('--gpu', action='store_true', - help="gpu") - parser.add_argument('--npu', action='store_true', - help="npu") - parser.add_argument('--amp', action='store_true', - help="amp") - parser.add_argument('--addr', default='127.0.0.1', - type=str, help='master addr') - parser.add_argument('--ignore_classifer', action='store_true', - help="ignore classifer layer weight when loading pretrained model") - args = parser.parse_args() - - cfg = get_default_config() - cfg.use_gpu = args.gpu - cfg.use_npu = args.npu - cfg.device_num = args.device_num - cfg.local_rank = args.local_rank - cfg.amp = args.amp - cfg.addr = args.addr - cfg.ignore_classifer = args.ignore_classifer - if args.config_file: - cfg.merge_from_file(args.config_file) - reset_config(cfg, args) - cfg.merge_from_list(args.opts) - set_random_seed(cfg.train.seed) - check_cfg(cfg) - - log_name = 'test.log' if cfg.test.evaluate else 'train.log' - log_name += time.strftime('-%Y-%m-%d-%H-%M-%S') - sys.stdout = Logger(osp.join(cfg.data.save_dir, log_name)) - - print('Show configuration\n{}\n'.format(cfg)) - print('Collecting env info ...') - print('** System info **\n{}\n'.format(collect_env_info())) - - if cfg.device_num == -1: - os.environ["CUDA_VISIBLE_DEVICES"] = '0' - os.environ['device_num'] = '-1' - else: - environ_str = '0' - for i in range(1, cfg.device_num): - environ_str = environ_str + ',%d' % i - os.environ["CUDA_VISIBLE_DEVICES"] = environ_str - os.environ['device_num'] = str(cfg.device_num) - - if cfg.use_gpu: - if cfg.device_num > 1: - torch.distributed.init_process_group(backend='nccl', init_method='env://') - torch.cuda.manual_seed_all(cfg.train.seed) - torch.cuda.set_device(cfg.local_rank) - os.environ['device'] = 'gpu' - torch.backends.cudnn.benchmark = True - - if cfg.use_npu: - os.environ['MASTER_ADDR'] = cfg.addr - os.environ['MASTER_PORT'] = '29688' - if cfg.device_num > 1: - torch.distributed.init_process_group(backend='hccl', rank=args.local_rank, world_size=args.device_num) - torch.npu.manual_seed_all(cfg.train.seed) - torch.npu.set_device(cfg.local_rank) - os.environ['device'] = 'npu' - - os.environ['batch_size'] = str(cfg.train.batch_size) - datamanager = build_datamanager(cfg) - - print('Building model: {}'.format(cfg.model.name)) - model = torchreid.models.build_model( - name=cfg.model.name, - num_classes=datamanager.num_train_pids, - loss=cfg.loss.name, - pretrained=cfg.model.pretrained, - use_gpu=cfg.use_gpu - ) - num_params, flops = compute_model_complexity( - model, (1, 3, cfg.data.height, cfg.data.width) - ) - print('Model complexity: params={:,} flops={:,}'.format(num_params, flops)) - - - if cfg.model.load_weights and check_isfile(cfg.model.load_weights): - load_pretrained_weights(model, cfg.model.load_weights, cfg.ignore_classifer) - - optimizer = torchreid.optim.build_optimizer(model, **optimizer_kwargs(cfg)) - - if cfg.use_gpu: - model = model.cuda() - elif cfg.use_npu: - model = model.npu() - - if cfg.amp: - if cfg.use_npu: - model, optimizer = amp.initialize(model, optimizer, opt_level="O2", loss_scale=128.0, combine_grad=True) - elif cfg.use_gpu: - model, optimizer = amp.initialize(model, optimizer, opt_level="O2", loss_scale=128.0) - - if cfg.device_num > 1: - model = nn.parallel.DistributedDataParallel(model, - device_ids=[cfg.local_rank], - output_device=cfg.local_rank, - find_unused_parameters=True, - broadcast_buffers=False - ) - - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, **lr_scheduler_kwargs(cfg) - ) - - if cfg.model.resume and check_isfile(cfg.model.resume): - cfg.train.start_epoch = resume_from_checkpoint( - cfg.model.resume, model, optimizer=optimizer, scheduler=scheduler - ) - - print( - 'Building {}-engine for {}-reid'.format(cfg.loss.name, cfg.data.type) - ) - engine = build_engine(cfg, datamanager, model, optimizer, scheduler) - if cfg.test.evaluate: - model.eval() - engine.test(**evaluate_kwargs(cfg)) - return - engine.run(**engine_run_kwargs(cfg)) - - -if __name__ == '__main__': - main() -- Gitee From 030738fe7ccec13317d3252e7364068f4b1d65ac Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:22 +0000 Subject: [PATCH 23/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/modelzoo=5Flevel.txt?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt | 3 --- 1 file changed, 3 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt diff --git a/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt b/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt deleted file mode 100644 index a3e2322b3a..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt +++ /dev/null @@ -1,3 +0,0 @@ -FuncStatus:OK -PerfStatus:OK -PrecisionStatus:OK -- Gitee From 79a6d787e5dfc49072e56e30b3366f4d35977a05 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:27 +0000 Subject: [PATCH 24/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/pthtar2onnx.py?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/pthtar2onnx.py | 57 ------------------- 1 file changed, 57 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py diff --git a/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py b/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py deleted file mode 100644 index 7f5e59dbcd..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py +++ /dev/null @@ -1,57 +0,0 @@ -# Copyright 2020 Huawei Technologies Co., Ltd -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# ============================================================================ - -import torch -import torch.onnx -import torchreid - -from collections import OrderedDict - - -def proc_node_module(checkpoint, AttrName): - new_state_dict = OrderedDict() - for k, v in checkpoint[AttrName].items(): - if(k[0:7] == "module."): - name = k[7:] - else: - name = k[0:] - new_state_dict[name] = v - return new_state_dict - - -def convert(): - checkpoint = torch.load('log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350', map_location='cpu') - checkpoint['state_dict'] = proc_node_module(checkpoint, 'state_dict') - model = torchreid.models.build_model( - name="osnet_x1_0", - num_classes=751, - loss="softmax", - pretrained=False, - use_gpu=False - ) - model.load_state_dict(checkpoint['state_dict']) - model.eval() - print(model) - - input_names = ["actual_input_1"] - output_names = ["output1"] - dummy_input = torch.randn(64, 3, 384, 128) - torch.onnx.export(model, dummy_input, "osnet.onnx", input_names=input_names, output_names=output_names, - opset_version=11) - print("export onnx done! save to osnet.onnx") - - -if __name__ == "__main__": - convert() -- Gitee From 9936614b31da4750bb29fbec4cf30378da462bef Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:33 +0000 Subject: [PATCH 25/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/requirements.txt?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/requirements.txt | 16 ---------------- 1 file changed, 16 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/requirements.txt diff --git a/PyTorch/contrib/cv/classification/OSNet/requirements.txt b/PyTorch/contrib/cv/classification/OSNet/requirements.txt deleted file mode 100644 index 8b35c0ebd9..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/requirements.txt +++ /dev/null @@ -1,16 +0,0 @@ -numpy -Cython -h5py -Pillow -six -scipy -opencv-python -matplotlib -tb-nightly -future -yacs -gdown -flake8 -yapf -isort -imageio \ No newline at end of file -- Gitee From e85e4d5990c612e51b1490d83d07ba9181a90ffc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:40 +0000 Subject: [PATCH 26/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/setup.py?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/setup.py | 104 ------------------ 1 file changed, 104 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/setup.py diff --git a/PyTorch/contrib/cv/classification/OSNet/setup.py b/PyTorch/contrib/cv/classification/OSNet/setup.py deleted file mode 100644 index 976b6c3949..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/setup.py +++ /dev/null @@ -1,104 +0,0 @@ -""" -BSD 3-Clause License - -Copyright (c) Soumith Chintala 2016, -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - -* Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - -* Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - -* Neither the name of the copyright holder nor the names of its - contributors may be used to endorse or promote products derived from - this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" -AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE -FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR -SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER -CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, -OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - - -Copyright 2020 Huawei Technologies Co., Ltd - -Licensed under the BSD 3-Clause License (the "License"); -you may not use this file except in compliance with the License. -You may obtain a copy of the License at - -https://spdx.org/licenses/BSD-3-Clause.html - -Unless required by applicable law or agreed to in writing, software -distributed under the License is distributed on an "AS IS" BASIS, -WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -See the License for the specific language governing permissions and -limitations under the License. -""" -import numpy as np -import os.path as osp -from setuptools import setup, find_packages -from distutils.extension import Extension -from Cython.Build import cythonize - - -def readme(): - with open('README.rst') as f: - content = f.read() - return content - - -def find_version(): - version_file = 'torchreid/__init__.py' - with open(version_file, 'r') as f: - exec(compile(f.read(), version_file, 'exec')) - return locals()['__version__'] - - -def numpy_include(): - try: - numpy_include = np.get_include() - except AttributeError: - numpy_include = np.get_numpy_include() - return numpy_include - - -ext_modules = [ - Extension( - 'torchreid.metrics.rank_cylib.rank_cy', - ['torchreid/metrics/rank_cylib/rank_cy.pyx'], - include_dirs=[numpy_include()], - ) -] - - -def get_requirements(filename='requirements.txt'): - here = osp.dirname(osp.realpath(__file__)) - with open(osp.join(here, filename), 'r') as f: - requires = [line.replace('\n', '') for line in f.readlines()] - return requires - - -setup( - name='torchreid', - version=find_version(), - description='A library for deep learning person re-ID in PyTorch', - author='Kaiyang Zhou', - license='MIT', - long_description=readme(), - url='https://github.com/KaiyangZhou/deep-person-reid', - packages=find_packages(), - install_requires=get_requirements(), - keywords=['Person Re-Identification', 'Deep Learning', 'Computer Vision'], - ext_modules=cythonize(ext_modules) -) -- Gitee From 0c59c4a350a9d82f16688887443d832d4f160915 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:31:57 +0000 Subject: [PATCH 27/31] init --- .../contrib/cv/classification/OSNet/.flake8 | 18 + .../cv/classification/OSNet/.gitignore | 146 ++++++++ .../cv/classification/OSNet/.isort.cfg | 10 + .../cv/classification/OSNet/.style.yapf | 7 + .../contrib/cv/classification/OSNet/LICENSE | 21 ++ .../contrib/cv/classification/OSNet/README.md | 72 ++++ .../cv/classification/OSNet/default_config.py | 272 +++++++++++++++ .../contrib/cv/classification/OSNet/demo.py | 187 +++++++++++ .../contrib/cv/classification/OSNet/main.py | 314 ++++++++++++++++++ .../classification/OSNet/modelzoo_level.txt | 3 + .../cv/classification/OSNet/pthtar2onnx.py | 57 ++++ .../cv/classification/OSNet/requirements.txt | 16 + .../contrib/cv/classification/OSNet/setup.py | 104 ++++++ 13 files changed, 1227 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/.flake8 create mode 100644 PyTorch/contrib/cv/classification/OSNet/.gitignore create mode 100644 PyTorch/contrib/cv/classification/OSNet/.isort.cfg create mode 100644 PyTorch/contrib/cv/classification/OSNet/.style.yapf create mode 100644 PyTorch/contrib/cv/classification/OSNet/LICENSE create mode 100644 PyTorch/contrib/cv/classification/OSNet/README.md create mode 100644 PyTorch/contrib/cv/classification/OSNet/default_config.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/demo.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/main.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py create mode 100644 PyTorch/contrib/cv/classification/OSNet/requirements.txt create mode 100644 PyTorch/contrib/cv/classification/OSNet/setup.py diff --git a/PyTorch/contrib/cv/classification/OSNet/.flake8 b/PyTorch/contrib/cv/classification/OSNet/.flake8 new file mode 100644 index 0000000000..4fc103cb10 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/.flake8 @@ -0,0 +1,18 @@ +[flake8] +ignore = + # At least two spaces before inline comment + E261, + # Line lengths are recommended to be no greater than 79 characters + E501, + # Missing whitespace around arithmetic operator + E226, + # Blank line contains whitespace + W293, + # Do not use bare 'except' + E722, + # Line break after binary operator + W504, + # isort found an import in the wrong position + I001 +max-line-length = 79 +exclude = __init__.py, build, torchreid/metrics/rank_cylib/ \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/.gitignore b/PyTorch/contrib/cv/classification/OSNet/.gitignore new file mode 100644 index 0000000000..7f6c2b6264 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/.gitignore @@ -0,0 +1,146 @@ +# Byte-compiled / optimized / DLL files +__pycache__/ +*.py[cod] +*$py.class + +nohup.out +inference/ +fusion_result.json +kernel_meta/ +cann_profiling/ +*.onnx +# C extensions +*.so + +# Distribution / packaging +.Python +build/ +develop-eggs/ +dist/ +downloads/ +eggs/ +.eggs/ +lib/ +lib64/ +parts/ +sdist/ +var/ +wheels/ +pip-wheel-metadata/ +share/python-wheels/ +*.egg-info/ +.installed.cfg +*.egg +MANIFEST + +# PyInstaller +# Usually these files are written by a python script from a template +# before PyInstaller builds the exe, so as to inject date/other infos into it. +*.manifest +*.spec + +# Installer logs +pip-log.txt +pip-delete-this-directory.txt + +# Unit test / coverage reports +htmlcov/ +.tox/ +.nox/ +.coverage +.coverage.* +.cache +nosetests.xml +coverage.xml +*.cover +.hypothesis/ +.pytest_cache/ + +# Translations +*.mo +*.pot + +# Django stuff: +*.log +local_settings.py +db.sqlite3 + +# Flask stuff: +instance/ +.webassets-cache + +# Scrapy stuff: +.scrapy + +# Sphinx documentation +docs/_build/ + +# PyBuilder +target/ + +# Jupyter Notebook +.ipynb_checkpoints + +# IPython +profile_default/ +ipython_config.py + +# pyenv +.python-version + +# pipenv +# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control. +# However, in case of collaboration, if having platform-specific dependencies or dependencies +# having no cross-platform support, pipenv may install dependencies that don't work, or not +# install all needed dependencies. +#Pipfile.lock + +# celery beat schedule file +celerybeat-schedule + +# SageMath parsed files +*.sage.py + +# Environments +.env +.venv +env/ +venv/ +ENV/ +env.bak/ +venv.bak/ + +# Spyder project settings +.spyderproject +.spyproject + +# Rope project settings +.ropeproject + +# mkdocs documentation +/site + +# mypy +.mypy_cache/ +.dmypy.json +dmypy.json + +# Pyre type checker +.pyre/ + +# Cython eval code +*.c +*.html + +# OS X +.DS_Store +.Spotlight-V100 +.Trashes +._* + +# ReID +reid-data/ +log/ +saved-models/ +model-zoo/ +debug* diff --git a/PyTorch/contrib/cv/classification/OSNet/.isort.cfg b/PyTorch/contrib/cv/classification/OSNet/.isort.cfg new file mode 100644 index 0000000000..8039326b5c --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/.isort.cfg @@ -0,0 +1,10 @@ +[isort] +line_length=79 +multi_line_output=3 +length_sort=true +known_standard_library=numpy,setuptools +known_myself=torchreid +known_third_party=matplotlib,cv2,torch,torchvision,PIL,yacs +no_lines_before=STDLIB,THIRDPARTY +sections=FUTURE,STDLIB,THIRDPARTY,myself,FIRSTPARTY,LOCALFOLDER +default_section=FIRSTPARTY \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/.style.yapf b/PyTorch/contrib/cv/classification/OSNet/.style.yapf new file mode 100644 index 0000000000..29d8e52cc8 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/.style.yapf @@ -0,0 +1,7 @@ +[style] +BASED_ON_STYLE = pep8 +BLANK_LINE_BEFORE_NESTED_CLASS_OR_DEF = true +SPLIT_BEFORE_EXPRESSION_AFTER_OPENING_PAREN = true +DEDENT_CLOSING_BRACKETS = true +SPACES_BEFORE_COMMENT = 1 +ARITHMETIC_PRECEDENCE_INDICATION = true \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/LICENSE b/PyTorch/contrib/cv/classification/OSNet/LICENSE new file mode 100644 index 0000000000..d2bcb88271 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2018 Kaiyang Zhou + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/PyTorch/contrib/cv/classification/OSNet/README.md b/PyTorch/contrib/cv/classification/OSNet/README.md new file mode 100644 index 0000000000..2528236875 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/README.md @@ -0,0 +1,72 @@ +# OSNet + +This implements training of OSNet on the Market-1501 dataset, mainly modified from [KaiyangZhou/deep-person-reid](https://github.com/KaiyangZhou/deep-person-reid). + +## OSNet Detail + +As of the current date, Ascend-Pytorch is still inefficient for contiguous operations.Therefore, OSNet is re-implemented using semantics such as custom OP. + + +## Requirements + +- Install PyTorch ([pytorch.org](http://pytorch.org)) + + +- `pip install -r requirements.txt` + +- Install torchreid + + - ~~~python + python setup.py develop + ~~~ + +- Download the Market-1501 dataset from https://paperswithcode.com/dataset/market-1501 + + - ~~~shell + unzip Market-1501-v15.09.15.zip + ~~~ + +- Move Market-1501 dataset to 'reid-data' path + + - ~~~shell + mkdir path_to_osnet/reid-data/ + mv Market-1501-v15.09.15 path_to_osnet/reid-data/market1501 + ~~~ +## Training + +To train a model, run `main.py` with the desired model architecture and the path to the ImageNet dataset: + +```bash +# training 1p accuracy +bash test/train_full_1p.sh + +# training 1p performance +bash test/train_performance_1p.sh + +# training 8p accuracy +bash test/train_full_8p.sh + +# training 8p performance +bash test/train_performance_8p.sh + +# finetuning +bash test/train_finetune_1p.sh --data_path=real_data_path --weight=real_weight_path + +# Online inference demo +python demo.py +## 备注： 识别前后图片保存到 `inference/` 文件夹下 + +# To ONNX +python pthtar2onnx.py +``` + +## OSNet training result + + +| | mAP | AMP_Type | Epochs | FPS | +| :----: | :--: | :------: | :----: | :------: | +| 1p-GPU | - | O2 | 1 | 371.383 | +| 1p-NPU | - | O2 | 1 | 366.464 | +| 8p-GPU | 80.3 | O2 | 350 | 1045.535 | +| 8p-NPU | 80.2 | O2 | 350 | 1091.358 | + diff --git a/PyTorch/contrib/cv/classification/OSNet/default_config.py b/PyTorch/contrib/cv/classification/OSNet/default_config.py new file mode 100644 index 0000000000..a2e22fe44d --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/default_config.py @@ -0,0 +1,272 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +from yacs.config import CfgNode as CN + + +def get_default_config(): + cfg = CN() + + # model + cfg.model = CN() + cfg.model.name = 'resnet50' + cfg.model.pretrained = True # automatically load pretrained model weights if available + cfg.model.load_weights = '' # path to model weights + cfg.model.resume = '' # path to checkpoint for resume training + + # data + cfg.data = CN() + cfg.data.type = 'image' + cfg.data.root = 'reid-data' + cfg.data.sources = ['market1501'] + cfg.data.targets = ['market1501'] + cfg.data.workers = 4 # number of data loading workers + cfg.data.split_id = 0 # split index + cfg.data.height = 256 # image height + cfg.data.width = 128 # image width + cfg.data.combineall = False # combine train, query and gallery for training + cfg.data.transforms = ['random_flip'] # data augmentation + cfg.data.k_tfm = 1 # number of times to apply augmentation to an image independently + cfg.data.norm_mean = [0.485, 0.456, 0.406] # default is imagenet mean + cfg.data.norm_std = [0.229, 0.224, 0.225] # default is imagenet std + cfg.data.save_dir = 'log' # path to save log + cfg.data.load_train_targets = False # load training set from target dataset + + # specific datasets + cfg.market1501 = CN() + cfg.market1501.use_500k_distractors = False # add 500k distractors to the gallery set for market1501 + cfg.cuhk03 = CN() + cfg.cuhk03.labeled_images = False # use labeled images, if False, use detected images + cfg.cuhk03.classic_split = False # use classic split by Li et al. CVPR14 + cfg.cuhk03.use_metric_cuhk03 = False # use cuhk03's metric for evaluation + + # sampler + cfg.sampler = CN() + cfg.sampler.train_sampler = 'RandomSampler' # sampler for source train loader + cfg.sampler.train_sampler_t = 'RandomSampler' # sampler for target train loader + cfg.sampler.num_instances = 4 # number of instances per identity for RandomIdentitySampler + cfg.sampler.num_cams = 1 # number of cameras to sample in a batch (for RandomDomainSampler) + cfg.sampler.num_datasets = 1 # number of datasets to sample in a batch (for RandomDatasetSampler) + + # video reid setting + cfg.video = CN() + cfg.video.seq_len = 15 # number of images to sample in a tracklet + cfg.video.sample_method = 'evenly' # how to sample images from a tracklet + cfg.video.pooling_method = 'avg' # how to pool features over a tracklet + + # train + cfg.train = CN() + cfg.train.optim = 'adam' + cfg.train.lr = 0.0003 + cfg.train.weight_decay = 5e-4 + cfg.train.max_epoch = 60 + cfg.train.start_epoch = 0 + cfg.train.batch_size = 32 + cfg.train.fixbase_epoch = 0 # number of epochs to fix base layers + cfg.train.open_layers = [ + 'classifier' + ] # layers for training while keeping others frozen + cfg.train.staged_lr = False # set different lr to different layers + cfg.train.new_layers = ['classifier'] # newly added layers with default lr + cfg.train.base_lr_mult = 0.1 # learning rate multiplier for base layers + cfg.train.lr_scheduler = 'single_step' + cfg.train.stepsize = [20] # stepsize to decay learning rate + cfg.train.gamma = 0.1 # learning rate decay multiplier + cfg.train.print_freq = 20 # print frequency + cfg.train.seed = 1 # random seed + + # optimizer + cfg.sgd = CN() + cfg.sgd.momentum = 0.9 # momentum factor for sgd and rmsprop + cfg.sgd.dampening = 0. # dampening for momentum + cfg.sgd.nesterov = False # Nesterov momentum + cfg.rmsprop = CN() + cfg.rmsprop.alpha = 0.99 # smoothing constant + cfg.adam = CN() + cfg.adam.beta1 = 0.9 # exponential decay rate for first moment + cfg.adam.beta2 = 0.999 # exponential decay rate for second moment + + # loss + cfg.loss = CN() + cfg.loss.name = 'softmax' + cfg.loss.softmax = CN() + cfg.loss.softmax.label_smooth = True # use label smoothing regularizer + cfg.loss.triplet = CN() + cfg.loss.triplet.margin = 0.3 # distance margin + cfg.loss.triplet.weight_t = 1. # weight to balance hard triplet loss + cfg.loss.triplet.weight_x = 0. # weight to balance cross entropy loss + + # test + cfg.test = CN() + cfg.test.batch_size = 100 + cfg.test.dist_metric = 'euclidean' # distance metric, ['euclidean', 'cosine'] + cfg.test.normalize_feature = False # normalize feature vectors before computing distance + cfg.test.ranks = [1, 5, 10, 20] # cmc ranks + cfg.test.evaluate = False # test only + cfg.test.eval_freq = -1 # evaluation frequency (-1 means to only test after training) + cfg.test.start_eval = 0 # start to evaluate after a specific epoch + cfg.test.rerank = False # use person re-ranking + cfg.test.visrank = False # visualize ranked results (only available when cfg.test.evaluate=True) + cfg.test.visrank_topk = 10 # top-k ranks to visualize + + return cfg + + +def imagedata_kwargs(cfg): + return { + 'root': cfg.data.root, + 'sources': cfg.data.sources, + 'targets': cfg.data.targets, + 'height': cfg.data.height, + 'width': cfg.data.width, + 'transforms': cfg.data.transforms, + 'k_tfm': cfg.data.k_tfm, + 'norm_mean': cfg.data.norm_mean, + 'norm_std': cfg.data.norm_std, + 'use_gpu': cfg.use_gpu, + 'split_id': cfg.data.split_id, + 'combineall': cfg.data.combineall, + 'load_train_targets': cfg.data.load_train_targets, + 'batch_size_train': cfg.train.batch_size, + 'batch_size_test': cfg.test.batch_size, + 'workers': cfg.data.workers, + 'num_instances': cfg.sampler.num_instances, + 'num_cams': cfg.sampler.num_cams, + 'num_datasets': cfg.sampler.num_datasets, + 'train_sampler': cfg.sampler.train_sampler, + 'train_sampler_t': cfg.sampler.train_sampler_t, + # image dataset specific + 'cuhk03_labeled': cfg.cuhk03.labeled_images, + 'cuhk03_classic_split': cfg.cuhk03.classic_split, + 'market1501_500k': cfg.market1501.use_500k_distractors, + 'device_num': cfg.device_num, + } + + +def videodata_kwargs(cfg): + return { + 'root': cfg.data.root, + 'sources': cfg.data.sources, + 'targets': cfg.data.targets, + 'height': cfg.data.height, + 'width': cfg.data.width, + 'transforms': cfg.data.transforms, + 'norm_mean': cfg.data.norm_mean, + 'norm_std': cfg.data.norm_std, + 'use_gpu': cfg.use_gpu, + 'split_id': cfg.data.split_id, + 'combineall': cfg.data.combineall, + 'batch_size_train': cfg.train.batch_size, + 'batch_size_test': cfg.test.batch_size, + 'workers': cfg.data.workers, + 'num_instances': cfg.sampler.num_instances, + 'num_cams': cfg.sampler.num_cams, + 'num_datasets': cfg.sampler.num_datasets, + 'train_sampler': cfg.sampler.train_sampler, + # video dataset specific + 'seq_len': cfg.video.seq_len, + 'sample_method': cfg.video.sample_method + } + + +def optimizer_kwargs(cfg): + return { + 'optim': cfg.train.optim, + 'lr': cfg.train.lr, + 'weight_decay': cfg.train.weight_decay, + 'momentum': cfg.sgd.momentum, + 'sgd_dampening': cfg.sgd.dampening, + 'sgd_nesterov': cfg.sgd.nesterov, + 'rmsprop_alpha': cfg.rmsprop.alpha, + 'adam_beta1': cfg.adam.beta1, + 'adam_beta2': cfg.adam.beta2, + 'staged_lr': cfg.train.staged_lr, + 'new_layers': cfg.train.new_layers, + 'base_lr_mult': cfg.train.base_lr_mult + } + + +def lr_scheduler_kwargs(cfg): + return { + 'lr_scheduler': cfg.train.lr_scheduler, + 'stepsize': cfg.train.stepsize, + 'gamma': cfg.train.gamma, + 'max_epoch': cfg.train.max_epoch + } + + +def engine_run_kwargs(cfg): + return { + 'save_dir': cfg.data.save_dir, + 'max_epoch': cfg.train.max_epoch, + 'start_epoch': cfg.train.start_epoch, + 'fixbase_epoch': cfg.train.fixbase_epoch, + 'open_layers': cfg.train.open_layers, + 'start_eval': cfg.test.start_eval, + 'eval_freq': cfg.test.eval_freq, + 'test_only': cfg.test.evaluate, + 'print_freq': cfg.train.print_freq, + 'dist_metric': cfg.test.dist_metric, + 'normalize_feature': cfg.test.normalize_feature, + 'visrank': cfg.test.visrank, + 'visrank_topk': cfg.test.visrank_topk, + 'use_metric_cuhk03': cfg.cuhk03.use_metric_cuhk03, + 'ranks': cfg.test.ranks, + 'rerank': cfg.test.rerank + } + +def evaluate_kwargs(cfg): + return { + 'dist_metric': cfg.test.dist_metric, + 'normalize_feature': cfg.test.normalize_feature, + 'visrank': cfg.test.visrank, + 'visrank_topk': cfg.test.visrank_topk, + 'save_dir': cfg.data.save_dir, + 'use_metric_cuhk03': cfg.cuhk03.use_metric_cuhk03, + 'ranks': cfg.test.ranks, + 'rerank': cfg.test.rerank + } \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/demo.py b/PyTorch/contrib/cv/classification/OSNet/demo.py new file mode 100644 index 0000000000..deb7c533fe --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/demo.py @@ -0,0 +1,187 @@ +#!/usr/bin/env python3 +# Copyright 2020 Huawei Technologies Co., Ltd +# +# Licensed under the BSD 3-Clause License (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# https://spdx.org/licenses/BSD-3-Clause.html +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +import torch +import numpy as np +import os.path as osp +import os +import argparse +import torchreid +from torchreid.data.datasets.image.market1501 import Market1501 +from torchreid.utils import load_pretrained_weights +from torchreid import metrics + +if not os.path.exists("inference"): + os.makedirs("inference") +os.system('rm -f inference/*') + +parser = argparse.ArgumentParser( + formatter_class=argparse.ArgumentDefaultsHelpFormatter +) +# data +parser.add_argument('-d', '--data_path', type=str, default='./reid-data/') +parser.add_argument('--device', type=str, default='cpu') +parser.add_argument('--checkpoint', type=str, default='log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350') +parser.add_argument('--config_file', type=str, default='configs/osnet_x1_0_trained_from_scratch.yaml') + +args = parser.parse_args() + +os.environ['device'] = args.device + +def build_model(): + # Create model + model = torchreid.models.build_model( + name="osnet_x1_0", + num_classes=751, + loss="softmax", + pretrained=False, + use_gpu=False + ) + load_pretrained_weights(model, args.checkpoint) + if os.environ['device'] == 'npu': + model = model.to("npu:0") + elif os.environ['device'] == 'gpu': + model = model.to("cuda:0") + model.eval() + return model + + +def get_raw_data(): + name, root = "market1501", args.data_path + + param = { + "root": root, + 'sources': ['market1501'], + 'targets': ['market1501'], + 'height': 256, + 'width': 128, + 'transforms': ['random_flip', 'random_crop', 'random_patch'], + 'k_tfm': 1, + 'norm_mean': [0.485, 0.456, 0.406], + 'norm_std': [0.229, 0.224, 0.225], + 'use_gpu': False, + 'split_id': 0, + 'combineall': False, + 'load_train_targets': False, + 'batch_size_train': 32, + 'batch_size_test': 32, + 'workers': 4, + 'num_instances': 4, + 'num_cams': 1, + 'num_datasets': 1, + 'train_sampler': "RandomSampler", + 'train_sampler_t': "RandomSampler", + # image dataset specific + 'cuhk03_labeled': False, + 'cuhk03_classic_split': False, + 'market1501_500k': False, + } + + datamanager = torchreid.data.ImageDataManager(**param) + + query_loader = datamanager.test_loader['market1501']['query'] + gallery_loader = datamanager.test_loader['market1501']['gallery'] + + data = next(iter(query_loader)) + + fnames = data['impath'] + pids = data['pid'] + imgs = data['img'] + camids = data['camid'] + + img = imgs[24] + pid = pids[24] + camid = camids[24] + fname = fnames[24] + img = torch.unsqueeze(img, dim=0) + + return datamanager, gallery_loader, img, fname, pid + +def parse_data_for_eval(data): + fnames = data['impath'] + imgs = data['img'] + pids = data['pid'] + camids = data['camid'] + return fnames, imgs, pids, camids + +def feature_extraction(model, imgs): + if os.environ['device'] == 'gpu': + imgs = imgs.cuda() + elif os.environ['device'] == 'npu': + imgs = imgs.npu() + features = model(imgs) + features = features.cpu().clone() + return features + +def find_imgs_with_id(id, data_loader): + f_, pids_, camids_, f_names_ = [], [], [], [] + for batch_idx, data in enumerate(data_loader): + fnames, imgs, pids, camids = parse_data_for_eval(data) + for fname, img, pid, camid in zip(fnames, imgs, pids, camids): + if pid == id: + img = torch.unsqueeze(img, dim=0) + f_.append(img) + pids_.append(pid) + camids_.append(camid) + f_names_.append(fname) + f_ = torch.cat(f_, dim=0) + return f_, pids_, camids_, f_names_ + +def feature_extraction_single(model, imgs): + if os.environ['device'] == 'gpu': + imgs = imgs.cuda() + elif os.environ['device'] == 'npu': + imgs = imgs.npu() + features = model(imgs) + features = features.cpu().clone() + return features + +def save_image(fname): + img_name = osp.basename(fname) + command = "cp %s ./inference/%s" % (fname, img_name) + os.system(command) + + +if __name__ == '__main__': + data_path = args.data_path + print("load dataset") + datamanager, gallery_loader, img, fname, pid = get_raw_data() + print("find a img in gallery with id %d" % pid) + imgs_gallery, pids, camids, f_names = find_imgs_with_id(pid.item(), gallery_loader) + + print("build model") + model = build_model() + print("extract img feature...") + img_feature = feature_extraction_single(model, img) + print("extract gallery feature...") + # gallery_feature = feature_extraction_single(model, img_gallery) + gallery_feature = feature_extraction(model, imgs_gallery) + + dist_metric = "euclidean" + print( + 'Computing distance matrix with metric={} ...'.format(dist_metric) + ) + distmat = metrics.compute_distance_matrix(img_feature, gallery_feature, dist_metric) + distmat = distmat.cpu().detach().numpy() + m, n = distmat.shape + indices = np.argsort(distmat, axis=1) + index = indices[0][0] + fname_gallery = f_names[index] + + save_image(fname) + save_image(fname_gallery) + print("query img saved to ./inference/%s" % osp.basename(fname)) + print("gallery img saved to ./inference/%s" % osp.basename(fname_gallery)) + \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/main.py b/PyTorch/contrib/cv/classification/OSNet/main.py new file mode 100644 index 0000000000..e70d026957 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/main.py @@ -0,0 +1,314 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +import sys +import time +import os.path as osp +import argparse +import torch +if torch.__version__>= '1.8.1': + import torch_npu +import torch.nn as nn +import os +#import torch_npu + +import torchreid +from torchreid.utils import ( + Logger, check_isfile, set_random_seed, collect_env_info, + resume_from_checkpoint, load_pretrained_weights, compute_model_complexity +) + +from default_config import ( + imagedata_kwargs, optimizer_kwargs, videodata_kwargs, engine_run_kwargs, + get_default_config, lr_scheduler_kwargs, evaluate_kwargs +) + +from apex import amp + + +def build_datamanager(cfg): + if cfg.data.type == 'image': + return torchreid.data.ImageDataManager(**imagedata_kwargs(cfg)) + else: + return torchreid.data.VideoDataManager(**videodata_kwargs(cfg)) + + +def build_engine(cfg, datamanager, model, optimizer, scheduler): + if cfg.data.type == 'image': + if cfg.loss.name == 'softmax': + engine = torchreid.engine.ImageSoftmaxEngine( + datamanager, + model, + optimizer=optimizer, + scheduler=scheduler, + use_gpu=cfg.use_gpu, + use_npu=cfg.use_npu, + label_smooth=cfg.loss.softmax.label_smooth, + use_amp=cfg.amp + ) + + else: + engine = torchreid.engine.ImageTripletEngine( + datamanager, + model, + optimizer=optimizer, + margin=cfg.loss.triplet.margin, + weight_t=cfg.loss.triplet.weight_t, + weight_x=cfg.loss.triplet.weight_x, + scheduler=scheduler, + use_gpu=cfg.use_gpu, + label_smooth=cfg.loss.softmax.label_smooth + ) + + else: + if cfg.loss.name == 'softmax': + engine = torchreid.engine.VideoSoftmaxEngine( + datamanager, + model, + optimizer=optimizer, + scheduler=scheduler, + use_gpu=cfg.use_gpu, + label_smooth=cfg.loss.softmax.label_smooth, + pooling_method=cfg.video.pooling_method + ) + + else: + engine = torchreid.engine.VideoTripletEngine( + datamanager, + model, + optimizer=optimizer, + margin=cfg.loss.triplet.margin, + weight_t=cfg.loss.triplet.weight_t, + weight_x=cfg.loss.triplet.weight_x, + scheduler=scheduler, + use_gpu=cfg.use_gpu, + label_smooth=cfg.loss.softmax.label_smooth + ) + + return engine + + +def reset_config(cfg, args): + if args.root: + cfg.data.root = args.root + if args.sources: + cfg.data.sources = args.sources + if args.targets: + cfg.data.targets = args.targets + if args.transforms: + cfg.data.transforms = args.transforms + + +def check_cfg(cfg): + if cfg.loss.name == 'triplet' and cfg.loss.triplet.weight_x == 0: + assert cfg.train.fixbase_epoch == 0, \ + 'The output of classifier is not included in the computational graph' + + +def main(): + parser = argparse.ArgumentParser( + formatter_class=argparse.ArgumentDefaultsHelpFormatter + ) + parser.add_argument( + '--config-file', type=str, default='', help='path to config file' + ) + parser.add_argument( + '-s', + '--sources', + type=str, + nargs='+', + help='source datasets (delimited by space)' + ) + parser.add_argument( + '-t', + '--targets', + type=str, + nargs='+', + help='target datasets (delimited by space)' + ) + parser.add_argument( + '--transforms', type=str, nargs='+', help='data augmentation' + ) + parser.add_argument( + '--root', type=str, default='', help='path to data root' + ) + parser.add_argument( + 'opts', + default=None, + nargs=argparse.REMAINDER, + help='Modify config options using the command-line' + ) + + parser.add_argument("--local_rank", type=int, default=0) + parser.add_argument('--device_num', default=-1, type=int, + help='device_num') + parser.add_argument('--gpu', action='store_true', + help="gpu") + parser.add_argument('--npu', action='store_true', + help="npu") + parser.add_argument('--amp', action='store_true', + help="amp") + parser.add_argument('--addr', default='127.0.0.1', + type=str, help='master addr') + parser.add_argument('--ignore_classifer', action='store_true', + help="ignore classifer layer weight when loading pretrained model") + args = parser.parse_args() + + cfg = get_default_config() + cfg.use_gpu = args.gpu + cfg.use_npu = args.npu + cfg.device_num = args.device_num + cfg.local_rank = args.local_rank + cfg.amp = args.amp + cfg.addr = args.addr + cfg.ignore_classifer = args.ignore_classifer + if args.config_file: + cfg.merge_from_file(args.config_file) + reset_config(cfg, args) + cfg.merge_from_list(args.opts) + set_random_seed(cfg.train.seed) + check_cfg(cfg) + + log_name = 'test.log' if cfg.test.evaluate else 'train.log' + log_name += time.strftime('-%Y-%m-%d-%H-%M-%S') + sys.stdout = Logger(osp.join(cfg.data.save_dir, log_name)) + + print('Show configuration\n{}\n'.format(cfg)) + print('Collecting env info ...') + print('** System info **\n{}\n'.format(collect_env_info())) + + if cfg.device_num == -1: + os.environ["CUDA_VISIBLE_DEVICES"] = '0' + os.environ['device_num'] = '-1' + else: + environ_str = '0' + for i in range(1, cfg.device_num): + environ_str = environ_str + ',%d' % i + os.environ["CUDA_VISIBLE_DEVICES"] = environ_str + os.environ['device_num'] = str(cfg.device_num) + + if cfg.use_gpu: + if cfg.device_num > 1: + torch.distributed.init_process_group(backend='nccl', init_method='env://') + torch.cuda.manual_seed_all(cfg.train.seed) + torch.cuda.set_device(cfg.local_rank) + os.environ['device'] = 'gpu' + torch.backends.cudnn.benchmark = True + + if cfg.use_npu: + os.environ['MASTER_ADDR'] = cfg.addr + os.environ['MASTER_PORT'] = '29688' + if cfg.device_num > 1: + torch.distributed.init_process_group(backend='hccl', rank=args.local_rank, world_size=args.device_num) + torch.npu.manual_seed_all(cfg.train.seed) + torch.npu.set_device(cfg.local_rank) + os.environ['device'] = 'npu' + + os.environ['batch_size'] = str(cfg.train.batch_size) + datamanager = build_datamanager(cfg) + + print('Building model: {}'.format(cfg.model.name)) + model = torchreid.models.build_model( + name=cfg.model.name, + num_classes=datamanager.num_train_pids, + loss=cfg.loss.name, + pretrained=cfg.model.pretrained, + use_gpu=cfg.use_gpu + ) + num_params, flops = compute_model_complexity( + model, (1, 3, cfg.data.height, cfg.data.width) + ) + print('Model complexity: params={:,} flops={:,}'.format(num_params, flops)) + + + if cfg.model.load_weights and check_isfile(cfg.model.load_weights): + load_pretrained_weights(model, cfg.model.load_weights, cfg.ignore_classifer) + + optimizer = torchreid.optim.build_optimizer(model, **optimizer_kwargs(cfg)) + + if cfg.use_gpu: + model = model.cuda() + elif cfg.use_npu: + model = model.npu() + + if cfg.amp: + if cfg.use_npu: + #model, optimizer = amp.initialize(model, optimizer, opt_level="O2", loss_scale=128.0, combine_grad=True) + model, optimizer = amp.initialize(model, optimizer, opt_level="O2", loss_scale="dynamic", combine_grad=True) + elif cfg.use_gpu: + model, optimizer = amp.initialize(model, optimizer, opt_level="O2", loss_scale=128.0) + + if cfg.device_num > 1: + model = nn.parallel.DistributedDataParallel(model, + device_ids=[cfg.local_rank], + output_device=cfg.local_rank, + find_unused_parameters=True, + broadcast_buffers=False + ) + + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, **lr_scheduler_kwargs(cfg) + ) + + if cfg.model.resume and check_isfile(cfg.model.resume): + cfg.train.start_epoch = resume_from_checkpoint( + cfg.model.resume, model, optimizer=optimizer, scheduler=scheduler + ) + + print( + 'Building {}-engine for {}-reid'.format(cfg.loss.name, cfg.data.type) + ) + engine = build_engine(cfg, datamanager, model, optimizer, scheduler) + if cfg.test.evaluate: + model.eval() + engine.test(**evaluate_kwargs(cfg)) + return + engine.run(**engine_run_kwargs(cfg)) + + +if __name__ == '__main__': + main() diff --git a/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt b/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt new file mode 100644 index 0000000000..a3e2322b3a --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/modelzoo_level.txt @@ -0,0 +1,3 @@ +FuncStatus:OK +PerfStatus:OK +PrecisionStatus:OK diff --git a/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py b/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py new file mode 100644 index 0000000000..7f5e59dbcd --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/pthtar2onnx.py @@ -0,0 +1,57 @@ +# Copyright 2020 Huawei Technologies Co., Ltd +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# ============================================================================ + +import torch +import torch.onnx +import torchreid + +from collections import OrderedDict + + +def proc_node_module(checkpoint, AttrName): + new_state_dict = OrderedDict() + for k, v in checkpoint[AttrName].items(): + if(k[0:7] == "module."): + name = k[7:] + else: + name = k[0:] + new_state_dict[name] = v + return new_state_dict + + +def convert(): + checkpoint = torch.load('log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350', map_location='cpu') + checkpoint['state_dict'] = proc_node_module(checkpoint, 'state_dict') + model = torchreid.models.build_model( + name="osnet_x1_0", + num_classes=751, + loss="softmax", + pretrained=False, + use_gpu=False + ) + model.load_state_dict(checkpoint['state_dict']) + model.eval() + print(model) + + input_names = ["actual_input_1"] + output_names = ["output1"] + dummy_input = torch.randn(64, 3, 384, 128) + torch.onnx.export(model, dummy_input, "osnet.onnx", input_names=input_names, output_names=output_names, + opset_version=11) + print("export onnx done! save to osnet.onnx") + + +if __name__ == "__main__": + convert() diff --git a/PyTorch/contrib/cv/classification/OSNet/requirements.txt b/PyTorch/contrib/cv/classification/OSNet/requirements.txt new file mode 100644 index 0000000000..8b35c0ebd9 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/requirements.txt @@ -0,0 +1,16 @@ +numpy +Cython +h5py +Pillow +six +scipy +opencv-python +matplotlib +tb-nightly +future +yacs +gdown +flake8 +yapf +isort +imageio \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/setup.py b/PyTorch/contrib/cv/classification/OSNet/setup.py new file mode 100644 index 0000000000..976b6c3949 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/setup.py @@ -0,0 +1,104 @@ +""" +BSD 3-Clause License + +Copyright (c) Soumith Chintala 2016, +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +* Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived from + this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + + +Copyright 2020 Huawei Technologies Co., Ltd + +Licensed under the BSD 3-Clause License (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + +https://spdx.org/licenses/BSD-3-Clause.html + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +""" +import numpy as np +import os.path as osp +from setuptools import setup, find_packages +from distutils.extension import Extension +from Cython.Build import cythonize + + +def readme(): + with open('README.rst') as f: + content = f.read() + return content + + +def find_version(): + version_file = 'torchreid/__init__.py' + with open(version_file, 'r') as f: + exec(compile(f.read(), version_file, 'exec')) + return locals()['__version__'] + + +def numpy_include(): + try: + numpy_include = np.get_include() + except AttributeError: + numpy_include = np.get_numpy_include() + return numpy_include + + +ext_modules = [ + Extension( + 'torchreid.metrics.rank_cylib.rank_cy', + ['torchreid/metrics/rank_cylib/rank_cy.pyx'], + include_dirs=[numpy_include()], + ) +] + + +def get_requirements(filename='requirements.txt'): + here = osp.dirname(osp.realpath(__file__)) + with open(osp.join(here, filename), 'r') as f: + requires = [line.replace('\n', '') for line in f.readlines()] + return requires + + +setup( + name='torchreid', + version=find_version(), + description='A library for deep learning person re-ID in PyTorch', + author='Kaiyang Zhou', + license='MIT', + long_description=readme(), + url='https://github.com/KaiyangZhou/deep-person-reid', + packages=find_packages(), + install_requires=get_requirements(), + keywords=['Person Re-Identification', 'Deep Learning', 'Computer Vision'], + ext_modules=cythonize(ext_modules) +) -- Gitee From 48d43dcd04a1d2698b484ba0ab01fabd7e0e6a35 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:32:36 +0000 Subject: [PATCH 28/31] init --- .../cv/classification/OSNet/README.rst | 317 + .../classification/OSNet/fusion_result.json | 87617 ++++++++++++++++ .../contrib/cv/classification/OSNet/nohup.out | 22941 ++++ 3 files changed, 110875 insertions(+) create mode 100644 PyTorch/contrib/cv/classification/OSNet/README.rst create mode 100644 PyTorch/contrib/cv/classification/OSNet/fusion_result.json create mode 100644 PyTorch/contrib/cv/classification/OSNet/nohup.out diff --git a/PyTorch/contrib/cv/classification/OSNet/README.rst b/PyTorch/contrib/cv/classification/OSNet/README.rst new file mode 100644 index 0000000000..57be7a86ba --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/README.rst @@ -0,0 +1,317 @@ +Torchreid +=========== +Torchreid is a library for deep-learning person re-identification, written in `PyTorch `_ and developed for our ICCV'19 project, `Omni-Scale Feature Learning for Person Re-Identification `_. + +It features: + +- multi-GPU training +- support both image- and video-reid +- end-to-end training and evaluation +- incredibly easy preparation of reid datasets +- multi-dataset training +- cross-dataset evaluation +- standard protocol used by most research papers +- highly extensible (easy to add models, datasets, training methods, etc.) +- implementations of state-of-the-art deep reid models +- access to pretrained reid models +- advanced training techniques +- visualization tools (tensorboard, ranks, etc.) + + +Code: https://github.com/KaiyangZhou/deep-person-reid. + +Documentation: https://kaiyangzhou.github.io/deep-person-reid/. + +How-to instructions: https://kaiyangzhou.github.io/deep-person-reid/user_guide. + +Model zoo: https://kaiyangzhou.github.io/deep-person-reid/MODEL_ZOO. + +Tech report: https://arxiv.org/abs/1910.10093. + +You can find some research projects that are built on top of Torchreid `here `_. + + +What's new +------------ +- [Aug 2021] We have released the ImageNet-pretrained models of ``osnet_ain_x0_75``, ``osnet_ain_x0_5`` and ``osnet_ain_x0_25``. The pretraining setup follows `pycls `_. +- [Apr 2021] We have updated the appendix in the `TPAMI version of OSNet `_ to include results in the multi-source domain generalization setting. The trained models can be found in the `Model Zoo `_. +- [Apr 2021] We have added a script to automate the process of calculating average results over multiple splits. For more details please see ``tools/parse_test_res.py``. +- [Apr 2021] ``v1.4.0``: We added the person search dataset, `CUHK-SYSU `_. Please see the `documentation `_ regarding how to download the dataset (it contains cropped person images). +- [Apr 2021] All models in the model zoo have been moved to google drive. Please raise an issue if any model's performance is inconsistent with the numbers shown in the model zoo page (could be caused by wrong links). +- [Mar 2021] `OSNet `_ will appear in the TPAMI journal! Compared with the conference version, which focuses on discriminative feature learning using the omni-scale building block, this journal extension further considers generalizable feature learning by integrating `instance normalization layers `_ with the OSNet architecture. We hope this journal paper can motivate more future work to taclke the generalization issue in cross-dataset re-ID. +- [Mar 2021] Generalization across domains (datasets) in person re-ID is crucial in real-world applications, which is closely related to the topic of *domain generalization*. Interested in learning how the field of domain generalization has developed over the last decade? Check our recent survey in this topic at https://arxiv.org/abs/2103.02503, with coverage on the history, datasets, related problems, methodologies, potential directions, and so on (*methods designed for generalizable re-ID are also covered*!). +- [Feb 2021] ``v1.3.6`` Added `University-1652 `_, a new dataset for multi-view multi-source geo-localization (credit to `Zhedong Zheng `_). +- [Feb 2021] ``v1.3.5``: Now the `cython code `_ works on Windows (credit to `lablabla `_). +- [Jan 2021] Our recent work, `MixStyle `_ (mixing instance-level feature statistics of samples of different domains for improving domain generalization), has been accepted to ICLR'21. The code has been released at https://github.com/KaiyangZhou/mixstyle-release where the person re-ID part is based on Torchreid. +- [Jan 2021] A new evaluation metric called `mean Inverse Negative Penalty (mINP)` for person re-ID has been introduced in `Deep Learning for Person Re-identification: A Survey and Outlook (TPAMI 2021) `_. Their code can be accessed at ``_. +- [Aug 2020] ``v1.3.3``: Fixed bug in ``visrank`` (caused by not unpacking ``dsetid``). +- [Aug 2020] ``v1.3.2``: Added ``_junk_pids`` to ``grid`` and ``prid``. This avoids using mislabeled gallery images for training when setting ``combineall=True``. +- [Aug 2020] ``v1.3.0``: (1) Added ``dsetid`` to the existing 3-tuple data source, resulting in ``(impath, pid, camid, dsetid)``. This variable denotes the dataset ID and is useful when combining multiple datasets for training (as a dataset indicator). E.g., when combining ``market1501`` and ``cuhk03``, the former will be assigned ``dsetid=0`` while the latter will be assigned ``dsetid=1``. (2) Added ``RandomDatasetSampler``. Analogous to ``RandomDomainSampler``, ``RandomDatasetSampler`` samples a certain number of images (``batch_size // num_datasets``) from each of specified datasets (the amount is determined by ``num_datasets``). +- [Aug 2020] ``v1.2.6``: Added ``RandomDomainSampler`` (it samples ``num_cams`` cameras each with ``batch_size // num_cams`` images to form a mini-batch). +- [Jun 2020] ``v1.2.5``: (1) Dataloader's output from ``__getitem__`` has been changed from ``list`` to ``dict``. Previously, an element, e.g. image tensor, was fetched with ``imgs=data[0]``. Now it should be obtained by ``imgs=data['img']``. See this `commit `_ for detailed changes. (2) Added ``k_tfm`` as an option to image data loader, which allows data augmentation to be applied ``k_tfm`` times *independently* to an image. If ``k_tfm > 1``, ``imgs=data['img']`` returns a list with ``k_tfm`` image tensors. +- [May 2020] Added the person attribute recognition code used in `Omni-Scale Feature Learning for Person Re-Identification (ICCV'19) `_. See ``projects/attribute_recognition/``. +- [May 2020] ``v1.2.1``: Added a simple API for feature extraction (``torchreid/utils/feature_extractor.py``). See the `documentation `_ for the instruction. +- [Apr 2020] Code for reproducing the experiments of `deep mutual learning `_ in the `OSNet paper `__ (Supp. B) has been released at ``projects/DML``. +- [Apr 2020] Upgraded to ``v1.2.0``. The engine class has been made more model-agnostic to improve extensibility. See `Engine `_ and `ImageSoftmaxEngine `_ for more details. Credit to `Dassl.pytorch `_. +- [Dec 2019] Our `OSNet paper `_ has been updated, with additional experiments (in section B of the supplementary) showing some useful techniques for improving OSNet's performance in practice. +- [Nov 2019] ``ImageDataManager`` can load training data from target datasets by setting ``load_train_targets=True``, and the train-loader can be accessed with ``train_loader_t = datamanager.train_loader_t``. This feature is useful for domain adaptation research. + + +Installation +--------------- + +Make sure `conda `_ is installed. + + +.. code-block:: bash + + # cd to your preferred directory and clone this repo + git clone https://github.com/KaiyangZhou/deep-person-reid.git + + # create environment + cd deep-person-reid/ + conda create --name torchreid python=3.7 + conda activate torchreid + + # install dependencies + # make sure `which python` and `which pip` point to the correct path + pip install -r requirements.txt + + # install torch and torchvision (select the proper cuda version to suit your machine) + conda install pytorch torchvision cudatoolkit=9.0 -c pytorch + + # install torchreid (don't need to re-build it if you modify the source code) + python setup.py develop + + +Get started: 30 seconds to Torchreid +------------------------------------- +1. Import ``torchreid`` + +.. code-block:: python + + import torchreid + +2. Load data manager + +.. code-block:: python + + datamanager = torchreid.data.ImageDataManager( + root="reid-data", + sources="market1501", + targets="market1501", + height=256, + width=128, + batch_size_train=32, + batch_size_test=100, + transforms=["random_flip", "random_crop"] + ) + +3 Build model, optimizer and lr_scheduler + +.. code-block:: python + + model = torchreid.models.build_model( + name="resnet50", + num_classes=datamanager.num_train_pids, + loss="softmax", + pretrained=True + ) + + model = model.cuda() + + optimizer = torchreid.optim.build_optimizer( + model, + optim="adam", + lr=0.0003 + ) + + scheduler = torchreid.optim.build_lr_scheduler( + optimizer, + lr_scheduler="single_step", + stepsize=20 + ) + +4. Build engine + +.. code-block:: python + + engine = torchreid.engine.ImageSoftmaxEngine( + datamanager, + model, + optimizer=optimizer, + scheduler=scheduler, + label_smooth=True + ) + +5. Run training and test + +.. code-block:: python + + engine.run( + save_dir="log/resnet50", + max_epoch=60, + eval_freq=10, + print_freq=10, + test_only=False + ) + + +A unified interface +----------------------- +In "deep-person-reid/scripts/", we provide a unified interface to train and test a model. See "scripts/main.py" and "scripts/default_config.py" for more details. The folder "configs/" contains some predefined configs which you can use as a starting point. + +Below we provide an example to train and test `OSNet (Zhou et al. ICCV'19) `_. Assume :code:`PATH_TO_DATA` is the directory containing reid datasets. The environmental variable :code:`CUDA_VISIBLE_DEVICES` is omitted, which you need to specify if you have a pool of gpus and want to use a specific set of them. + +Conventional setting +^^^^^^^^^^^^^^^^^^^^^ + +To train OSNet on Market1501, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA + + +The config file sets Market1501 as the default dataset. If you wanna use DukeMTMC-reID, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + -s dukemtmcreid \ + -t dukemtmcreid \ + --transforms random_flip random_erase \ + --root $PATH_TO_DATA \ + data.save_dir log/osnet_x1_0_dukemtmcreid_softmax_cosinelr + +The code will automatically (download and) load the ImageNet pretrained weights. After the training is done, the model will be saved as "log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250". Under the same folder, you can find the `tensorboard `_ file. To visualize the learning curves using tensorboard, you can run :code:`tensorboard --logdir=log/osnet_x1_0_market1501_softmax_cosinelr` in the terminal and visit :code:`http://localhost:6006/` in your web browser. + +Evaluation is automatically performed at the end of training. To run the test again using the trained model, do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ + --root $PATH_TO_DATA \ + model.load_weights log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250 \ + test.evaluate True + + +Cross-domain setting +^^^^^^^^^^^^^^^^^^^^^ + +Suppose you wanna train OSNet on DukeMTMC-reID and test its performance on Market1501, you can do + +.. code-block:: bash + + python scripts/main.py \ + --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad.yaml \ + -s dukemtmcreid \ + -t market1501 \ + --transforms random_flip color_jitter \ + --root $PATH_TO_DATA + +Here we only test the cross-domain performance. However, if you also want to test the performance on the source dataset, i.e. DukeMTMC-reID, you can set :code:`-t dukemtmcreid market1501`, which will evaluate the model on the two datasets separately. + +Different from the same-domain setting, here we replace :code:`random_erase` with :code:`color_jitter`. This can improve the generalization performance on the unseen target dataset. + +Pretrained models are available in the `Model Zoo `_. + + +Datasets +-------- + +Image-reid datasets +^^^^^^^^^^^^^^^^^^^^^ +- `Market1501 `_ +- `CUHK03 `_ +- `DukeMTMC-reID `_ +- `MSMT17 `_ +- `VIPeR `_ +- `GRID `_ +- `CUHK01 `_ +- `SenseReID `_ +- `QMUL-iLIDS `_ +- `PRID `_ + +Geo-localization datasets +^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `University-1652 `_ + +Video-reid datasets +^^^^^^^^^^^^^^^^^^^^^^^ +- `MARS `_ +- `iLIDS-VID `_ +- `PRID2011 `_ +- `DukeMTMC-VideoReID `_ + + +Models +------- + +ImageNet classification models +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ +- `ResNet `_ +- `ResNeXt `_ +- `SENet `_ +- `DenseNet `_ +- `Inception-ResNet-V2 `_ +- `Inception-V4 `_ +- `Xception `_ +- `IBN-Net `_ + +Lightweight models +^^^^^^^^^^^^^^^^^^^ +- `NASNet `_ +- `MobileNetV2 `_ +- `ShuffleNet `_ +- `ShuffleNetV2 `_ +- `SqueezeNet `_ + +ReID-specific models +^^^^^^^^^^^^^^^^^^^^^^ +- `MuDeep `_ +- `ResNet-mid `_ +- `HACNN `_ +- `PCB `_ +- `MLFN `_ +- `OSNet `_ +- `OSNet-AIN `_ + + +Useful links +------------- +- `OSNet-IBN1-Lite (test-only code with lite docker container) `_ +- `Deep Learning for Person Re-identification: A Survey and Outlook `_ + + +Citation +--------- +If you use this code or the models in your research, please give credit to the following papers: + +.. code-block:: bash + + @article{torchreid, + title={Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch}, + author={Zhou, Kaiyang and Xiang, Tao}, + journal={arXiv preprint arXiv:1910.10093}, + year={2019} + } + + @inproceedings{zhou2019osnet, + title={Omni-Scale Feature Learning for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + booktitle={ICCV}, + year={2019} + } + + @article{zhou2021osnet, + title={Learning Generalisable Omni-Scale Representations for Person Re-Identification}, + author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, + journal={TPAMI}, + year={2021} + } diff --git a/PyTorch/contrib/cv/classification/OSNet/fusion_result.json b/PyTorch/contrib/cv/classification/OSNet/fusion_result.json new file mode 100644 index 0000000000..2f45d05310 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/fusion_result.json @@ -0,0 +1,87617 @@ +{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "0_0" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "100_100" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "101_101" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "102_102" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "103_103" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "104_104" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "105_105" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "106_106" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "107_107" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "108_108" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "109_109" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "10_10" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "110_110" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "111_111" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "112_112" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "113_113" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "114_114" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "115_115" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "116_116" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "117_117" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "118_118" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "119_119" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "11_11" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "120_120" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "121_121" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "122_122" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "123_123" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "124_124" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "125_125" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "126_126" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "127_127" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "128_128" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "129_129" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "12_12" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "130_130" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "131_131" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "132_132" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "133_133" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "9" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "134_134" +}{ + "graph_fusion": { + "ConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "SplitConvConcatFusionPass": { + "effect_times": "0", + "match_times": "3" + }, + "ZConcatDFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "135_135" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "136_136" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "137_137" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "138_138" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "139_139" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "13_13" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "140_140" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "141_141" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "142_142" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "143_143" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "144_144" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "145_145" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "146_146" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "147_147" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "148_148" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "149_149" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "14_14" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "150_150" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "151_151" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "152_152" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "153_153" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "154_154" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "155_155" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "156_156" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "157_157" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "158_158" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "159_159" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "15_15" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "160_160" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "161_161" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "162_162", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "163_163" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "164_164" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "165_165" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "166_166" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "167_167" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "168_168" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "169_169" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "16_16" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "170_170" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "171_171" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "172_172" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "173_173" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "174_174" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "175_175" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "176_176" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "177_177" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "178_178" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "179_179" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "17_17" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "180_180" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "181_181" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "182_182" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "183_183" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "184_184", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "185_185" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "186_186" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "187_187" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "188_188" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "189_189" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "18_18" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "190_190" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "191_191" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "192_192" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "193_193" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "194_194" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "195_195" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "196_196" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "197_197" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "198_198" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "199_199" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "19_19" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "1_1" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "200_200" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "201_201" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "202_202" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "203_203" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "204_204" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "205_205" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "206_206" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "207_207" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "208_208" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "209_209" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "20_20" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "210_210" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "211_211" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "212_212" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "213_213" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "214_214" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "215_215" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "216_216" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "217_217" +}{ + "graph_fusion": { + "LogSoftmaxFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "218_218" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "219_219" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "21_21" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "220_220" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "221_221" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "222_222" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "223_223" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "224_224" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "225_225" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "226_226" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "227_227" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "228_228" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "229_229" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "22_22" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "230_230" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "231_231" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "232_232" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "233_233" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "234_234" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "235_235" +}{ + "graph_fusion": { + "RealDiv2MulsFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "236_236" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "237_237" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "238_238" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "239_239" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "23_23" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "240_240" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "241_241" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "242_242" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "243_243" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "244_244" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "245_245" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "3" + } + }, + "session_and_graph_id": "246_246" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "247_247" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "248_248" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "249_249" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "24_24" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "250_250" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "251_251" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "252_252" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "253_253" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "254_254" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "255_255" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "256_256" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "257_257" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "258_258" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "259_259" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "25_25" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "260_260" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "261_261" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "262_262" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "263_263" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "264_264" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "265_265" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "266_266" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "267_267" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "268_268" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "269_269" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "26_26" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "270_270" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "271_271" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "272_272" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "273_273" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "274_274" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "275_275" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "276_276" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "277_277" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "278_278" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "279_279" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "27_27" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "280_280" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "281_281" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "282_282" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "283_283" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "284_284" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "285_285" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "286_286" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "287_287" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "288_288" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "289_289" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "28_28" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "290_290" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "291_291" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "292_292" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "293_293" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "294_294" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "295_295" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "296_296" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "297_297" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "298_298" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "299_299" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "29_29" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "2_2" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "300_300" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "301_301" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "302_302" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "303_303" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "304_304" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "305_305" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "306_306" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "307_307" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "308_308" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "309_309" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "30_30" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "310_310" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "311_311" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "312_312" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "313_313" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "314_314" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "315_315" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "316_316" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "317_317" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "318_318" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "319_319" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "31_31" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "320_320" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "321_321" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "322_322" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "323_323" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "324_324" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "325_325" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "326_326" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "327_327" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "328_328" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "329_329" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "32_32" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "330_330" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "331_331" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "332_332" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "333_333" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "334_334" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "335_335" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "336_336" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "337_337" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "338_338" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "339_339" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "33_33" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "340_340" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "341_341" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "342_342" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "343_343" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "344_344" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "345_345" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "346_346" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "347_347" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "348_348" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "349_349" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "34_34" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "350_350" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "351_351" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "352_352" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "353_353" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "354_354" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "355_355" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "356_356" +}{ + "graph_fusion": { + "AvgPoolV2GradFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "357_357" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "358_358" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "359_359" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "35_35" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "360_360" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "361_361" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "362_362" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "363_363" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "364_364" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "365_365" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "366_366" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "367_367" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "368_368" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "369_369" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "36_36" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "370_370" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "371_371" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "372_372" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "373_373" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "374_374" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "375_375" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "376_376" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "377_377" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "378_378" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "379_379" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "37_37" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "380_380" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "381_381" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "382_382" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "383_383" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "384_384" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "385_385" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "386_386" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "387_387" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "388_388" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "389_389" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "38_38" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "390_390" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "391_391" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "392_392" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "393_393" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + }, + "TransposedUpdateFusionPass": { + "effect_times": "2", + "match_times": "2" + } + }, + "session_and_graph_id": "394_394" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "395_395" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "396_396" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "397_397" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "398_398" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpInputDilationFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "399_399" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "39_39" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "3_3" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "400_400" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "401_401" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "402_402" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "403_403" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "404_404" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "405_405" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "Conv2DbpFilterMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "406_406" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "407_407" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "408_408" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "409_409" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "40_40" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "410_410" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "411_411" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "412_412" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "413_413" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "414_414" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "415_415" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "416_416" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "417_417" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "418_418" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "419_419" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "41_41" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "420_420" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "421_421" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "422_422" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "423_423" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "424_424" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "425_425" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "426_426" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "427_427" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "428_428" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "429_429" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "42_42" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "430_430" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "431_431" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "432_432" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "433_433" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "434_434" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "435_435" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "436_436" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "437_437" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "438_438" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "439_439" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "43_43" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "440_440" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "441_441" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "442_442" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "443_443" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "444_444" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "445_445" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "446_446" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "447_447" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "448_448" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "449_449" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "44_44" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "450_450" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "451_451" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "452_452" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "453_453" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "454_454" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "455_455" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "456_456" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "457_457" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "458_458" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "459_459" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "45_45" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "460_460" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "461_461" +}{ + "graph_fusion": { + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "462_462" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "463_463" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "464_464" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "465_465" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "466_466" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TopKFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "467_467" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "468_468" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "469_469" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "46_46" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "470_470" +}{ + "graph_fusion": { + "AReduceSumFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrReduceSumFusion": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "471_471" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "472_472" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "473_473" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "474_474" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "475_475" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "476_476" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "477_477" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "478_478" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "479_479" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "47_47" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "480_480" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "481_481" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "482_482" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "483_483" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "484_484" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "485_485" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "486_486" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "487_487" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "488_488" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "489_489" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "48_48" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "490_490" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "491_491" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "492_492" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "493_493" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "494_494" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "495_495" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "496_496" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "497_497" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "498_498" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "499_499" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "49_49" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "4_4" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "500_500" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "501_501" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "502_502" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "503_503" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "504_504" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "505_505" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "506_506" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "507_507" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "508_508" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "509_509" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "50_50" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "510_510" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "511_511" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "512_512" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "513_513" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "514_514" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "515_515" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "516_516" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "517_517" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "518_518" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "519_519" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "51_51" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "520_520" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "521_521" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "522_522" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "523_523" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "524_524" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "525_525" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "526_526" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "527_527" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "528_528" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "529_529" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "52_52" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "530_530" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "531_531" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "532_532" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "533_533" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "534_534" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "535_535" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "536_536" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "537_537" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "538_538" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "539_539" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "53_53" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "540_540" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "541_541" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "542_542" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "543_543" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "544_544" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "545_545" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "546_546" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "547_547" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "548_548" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "549_549" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "54_54" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "550_550" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "551_551" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "552_552" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "553_553" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "554_554" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "555_555" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "556_556" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "557_557" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "558_558" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "559_559" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "55_55" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "560_560" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "561_561" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "562_562" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "563_563" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "564_564" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "565_565" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "566_566" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "567_567" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "568_568" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "569_569" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "56_56" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "570_570" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "571_571" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "572_572" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "573_573" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "574_574" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "575_575" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "576_576", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "577_577" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "578_578" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "579_579" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "57_57" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "580_580" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "581_581" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "582_582" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "583_583" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "584_584" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "585_585" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "586_586" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "587_587" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "588_588" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "589_589" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "58_58" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "590_590" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "591_591" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "592_592" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "593_593" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "594_594" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "595_595" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "596_596", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "597_597" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "598_598" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "599_599" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "59_59" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "5_5" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "600_600" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "601_601" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "602_602" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "603_603" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "604_604" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "605_605" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "606_606" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "607_607" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "608_608" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "609_609" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "60_60" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "610_610" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "611_611" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "612_612" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "613_613" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "614_614" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "615_615" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "616_616" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "617_617" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "618_618" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "619_619" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "61_61" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "620_620" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "621_621" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "622_622" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "623_623" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "624_624" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "625_625" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "626_626" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "627_627" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "628_628" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "629_629" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "62_62" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "630_630" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "631_631" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "632_632" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "633_633" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "634_634" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "635_635" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "636_636" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "637_637" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "638_638" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "639_639" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "63_63" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "640_640" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "641_641" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "642_642" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "643_643" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "644_644" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "645_645" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "646_646" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "647_647", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "648_648" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "649_649" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "64_64" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "650_650" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "651_651" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "652_652" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "653_653" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "654_654" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "655_655" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "656_656" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "657_657" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "658_658" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "659_659" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "65_65" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "660_660" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "661_661" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "662_662" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "663_663" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "664_664" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "665_665" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "666_666" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "667_667", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "668_668" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "669_669" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "66_66" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "670_670" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "671_671" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "672_672" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "673_673" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "674_674" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "675_675" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "676_676" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "677_677" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "678_678" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "679_679" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "67_67" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "680_680" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "681_681" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "682_682" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "683_683" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "684_684" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "685_685" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "686_686" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "687_687" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "688_688" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "689_689" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "68_68" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "690_690" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "691_691" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "692_692" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "693_693" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "694_694" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "695_695" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "696_696" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "697_697" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "698_698" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "699_699" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "69_69" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "6_6" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "700_700" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "701_701" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "702_702" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "703_703" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "704_704" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "705_705" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "706_706" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "707_707" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "708_708" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "709_709" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "70_70" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "710_710" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "711_711" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "712_712" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "713_713" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "714_714" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "715_715" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "716_716" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "717_717" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "718_718", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "719_719" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "71_71" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "720_720" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "721_721" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "722_722" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "723_723" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "724_724" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "725_725" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "726_726" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "727_727" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "728_728" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "729_729" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "72_72" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "730_730" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "731_731" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "732_732" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "733_733" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "734_734" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "735_735" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "736_736" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "737_737" +}{ + "graph_fusion": { + "AvgPoolV2FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "738_738", + "ub_fusion": { + "TbeCommonRules0FusionPass": { + "effect_times": "1", + "match_times": "1" + } + } +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "739_739" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "73_73" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "740_740" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "741_741" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "742_742" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "TransposedUpdateFusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "743_743" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "744_744" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "745_745" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "746_746" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "747_747" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "748_748" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "749_749" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "74_74" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "750_750" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "751_751" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "752_752" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "753_753" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "754_754" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "755_755" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "756_756" +}{ + "graph_fusion": { + "ConvToFullyConnectionFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConvWeightCompressFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "757_757" +}{ + "graph_fusion": { + "AReduceMeanFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ConstToAttrPass": { + "effect_times": "1", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "758_758" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "759_759" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "75_75" +}{ + "graph_fusion": { + "MulSquareFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "760_760" +}{ + "graph_fusion": { + "BatchMatMulFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEAPREQUANTFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "FIXPIPEFUSIONPASS": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "761_761" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "762_762" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "763_763" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "2" + } + }, + "session_and_graph_id": "764_764" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "765_765" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "766_766" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "76_76" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "77_77" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "78_78" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "79_79" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "7_7" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "80_80" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "81_81" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "82_82" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "83_83" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "84_84" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "85_85" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "86_86" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "87_87" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "88_88" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "89_89" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "8_8" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "90_90" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "91_91" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "92_92" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "93_93" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "94_94" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "95_95" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "96_96" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "97_97" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "98_98" +}{ + "graph_fusion": { + "CastRemoveFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "ForceFp16CastFusionPass": { + "effect_times": "0", + "match_times": "1" + }, + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + } + }, + "session_and_graph_id": "99_99" +}{ + "graph_fusion": { + "RefreshInt64ToInt32FusionPass": { + "effect_times": "1", + "match_times": "1" + }, + "TransdataCastFusionPass": { + "effect_times": "0", + "match_times": "1" + } + }, + "session_and_graph_id": "9_9" +} \ No newline at end of file diff --git a/PyTorch/contrib/cv/classification/OSNet/nohup.out b/PyTorch/contrib/cv/classification/OSNet/nohup.out new file mode 100644 index 0000000000..0491b2f677 --- /dev/null +++ b/PyTorch/contrib/cv/classification/OSNet/nohup.out @@ -0,0 +1,22941 @@ +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 0 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Traceback (most recent call last): + File "main.py", line 310, in + main() + File "main.py", line 243, in main + torch.distributed.init_process_group(backend='hccl', rank=args.local_rank, world_size=args.device_num) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 420, in init_process_group + store, rank, world_size = next(rendezvous_iterator) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/distributed/rendezvous.py", line 172, in _env_rendezvous_handler + store = TCPStore(master_addr, master_port, world_size, start_daemon, timeout) +RuntimeError: Address already in use +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 6 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 7 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 0 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 1 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 4 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 5 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 2 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 3 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 6 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 7 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 1 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 3 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 4 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 5 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 2 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 1 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.5.0+ascend.post3 +Is debug build: No +CUDA used to build PyTorch: None + +OS: CentOS Linux 7 (AltArch) +GCC version: (GCC) 7.3.0 +CMake version: version 3.18.6 + +Python version: 3.7 +Is CUDA available: No +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.5.0+ascend.post3.20210930 +[conda] torch 1.5.0+ascend.post3.20210930 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. + 'Cython evaluation (very fast so highly recommended) is ' +/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". + 'The current data structure is deprecated. Please ' +Use npu fused optimizer +Use npu fused optimizer +Use npu fused optimizer +Use npu fused optimizer +Use npu fused optimizer +Use npu fused optimizer +Use npu fused optimizer +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +THPModule_npu_shutdown success. +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +THPModule_npu_shutdown success. +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +THPModule_npu_shutdown success. +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +Traceback (most recent call last): + File "main.py", line 310, in + if __name__ == '__main__': + File "main.py", line 286, in main + find_unused_parameters=True, + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ + self.broadcast_bucket_size) + File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced + dist._broadcast_coalesced(self.process_group, tensors, buffer_size) +RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. + +Defaults for this optimization level are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : dynamic +combine_grad : None +check_combined_tensors : None +Processing user overrides (additional kwargs that are not None)... +After processing overrides, optimization options are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : 128.0 +combine_grad : True +check_combined_tensors : None +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +group num: 1 +epoch: [1/1][20/50] time 0.286 (0.294) data 0.000 (0.020) eta 0:00:08 loss 6.5586 (6.5816) acc 0.0000 (0.9375) lr 0.260000 +epoch: [1/1][40/50] time 0.285 (0.295) data 0.000 (0.010) eta 0:00:02 loss 6.0959 (6.4039) acc 0.0000 (1.0156) lr 0.260000 +FPS@all 866.618, TIME@all 0.295 +Elapsed 0:07:53 +FPS@all 866.618, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.285 (0.294) data 0.000 (0.015) eta 1:25:43 loss 6.6704 (6.6383) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 6.1446 (6.4308) acc 3.1250 (0.7812) lr 0.260000 +FPS@all 866.705, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.286 (0.294) data 0.000 (0.017) eta 1:25:42 loss 6.6813 (6.5624) acc 0.0000 (0.7812) lr 0.260000 +epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.009) eta 1:25:48 loss 5.7973 (6.3714) acc 3.1250 (1.3281) lr 0.260000 +FPS@all 866.715, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.284 (0.295) data 0.000 (0.016) eta 1:25:53 loss 6.2679 (6.5310) acc 6.2500 (1.4062) lr 0.260000 +epoch: [1/350][40/50] time 0.277 (0.295) data 0.000 (0.008) eta 1:25:48 loss 5.9363 (6.3801) acc 9.3750 (1.2500) lr 0.260000 +FPS@all 866.823, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.286 (0.294) data 0.000 (0.016) eta 1:25:42 loss 6.3082 (6.5779) acc 0.0000 (0.0000) lr 0.260000 +epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 6.0846 (6.4162) acc 3.1250 (0.6250) lr 0.260000 +FPS@all 866.665, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.284 (0.295) data 0.001 (0.012) eta 1:25:58 loss 6.7670 (6.5713) acc 3.1250 (0.4688) lr 0.260000 +epoch: [1/350][40/50] time 0.304 (0.296) data 0.000 (0.006) eta 1:26:06 loss 6.0424 (6.3826) acc 0.0000 (1.1719) lr 0.260000 +FPS@all 866.506, TIME@all 0.295 +group num: 1 +epoch: [1/350][20/50] time 0.285 (0.294) data 0.000 (0.016) eta 1:25:42 loss 6.5851 (6.6007) acc 0.0000 (0.9375) lr 0.260000 +epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 5.9063 (6.4240) acc 3.1250 (1.2500) lr 0.260000 +FPS@all 866.753, TIME@all 0.295 +group num: 1 +epoch: [1/1][20/50] time 0.236 (0.294) data 0.000 (0.015) eta 0:00:08 loss 6.5861 (6.5515) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/1][40/50] time 0.274 (0.294) data 0.001 (0.008) eta 0:00:02 loss 5.7942 (6.3841) acc 3.1250 (1.5625) lr 0.260000 +FPS@all 867.641, TIME@all 0.295 +Elapsed 0:07:45 +FPS@all 867.641, TIME@all 0.295 +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +terminate called after throwing an instance of 'std::runtime_error' + what(): AllReduce error in:/usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/ProcessGroupHCCL.cpp: 114 +/root/archiconda3/envs/ych/lib/python3.7/multiprocessing/semaphore_tracker.py:144: UserWarning: semaphore_tracker: There appear to be 91 leaked semaphores to clean up at shutdown + len(cache)) +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Traceback (most recent call last): + File "main.py", line 55, in + import torch_npu +ModuleNotFoundError: No module named 'torch_npu' +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 6 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 0 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 2 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 1 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 3 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 4 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 5 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 7 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. + +Defaults for this optimization level are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : dynamic +combine_grad : None +combine_ddp : None +ddp_replica_count : 4 +check_combined_tensors : None +user_cast_preferred : None +Processing user overrides (additional kwargs that are not None)... +After processing overrides, optimization options are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : 128.0 +combine_grad : True +combine_ddp : None +ddp_replica_count : 4 +check_combined_tensors : None +user_cast_preferred : None +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.020) eta 1:28:17 loss 6.4503 (6.6220) acc 0.0000 (0.9375) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:18 loss 6.1957 (6.4520) acc 0.0000 (1.2500) lr 0.260000 +FPS@all 844.099, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.021) eta 1:28:17 loss 6.4964 (6.6583) acc 0.0000 (0.1562) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:28:18 loss 6.1912 (6.4751) acc 3.1250 (0.7031) lr 0.260000 +FPS@all 844.086, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.021) eta 1:28:17 loss 6.5021 (6.5814) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:28:18 loss 5.8930 (6.4415) acc 9.3750 (1.1719) lr 0.260000 +FPS@all 844.105, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.019) eta 1:28:16 loss 6.6479 (6.6206) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.001 (0.010) eta 1:28:18 loss 6.2197 (6.4612) acc 0.0000 (1.0938) lr 0.260000 +FPS@all 844.124, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.020) eta 1:28:17 loss 6.7775 (6.5672) acc 0.0000 (0.6250) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:18 loss 6.0981 (6.4246) acc 3.1250 (1.1719) lr 0.260000 +FPS@all 844.117, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.001 (0.020) eta 1:28:15 loss 6.6308 (6.5626) acc 0.0000 (0.1562) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:17 loss 6.0067 (6.4354) acc 3.1250 (0.9375) lr 0.260000 +FPS@all 844.289, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.305 (0.303) data 0.001 (0.020) eta 1:28:15 loss 6.5050 (6.6098) acc 0.0000 (0.7812) lr 0.260000 +epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:17 loss 6.1065 (6.4767) acc 3.1250 (0.7812) lr 0.260000 +FPS@all 844.264, TIME@all 0.303 +group num: 1 +epoch: [1/350][20/50] time 0.304 (0.303) data 0.000 (0.020) eta 1:28:16 loss 6.3586 (6.5710) acc 0.0000 (0.6250) lr 0.260000 +epoch: [1/350][40/50] time 0.304 (0.303) data 0.000 (0.010) eta 1:28:18 loss 5.9643 (6.4363) acc 6.2500 (0.8594) lr 0.260000 +FPS@all 844.485, TIME@all 0.303 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:43 loss 5.6904 (5.6815) acc 0.0000 (3.1250) lr 0.260000 +epoch: [2/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.6236 (5.6524) acc 6.2500 (3.3594) lr 0.260000 +FPS@all 841.331, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.9472 (5.6783) acc 3.1250 (4.3750) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:27 loss 6.0357 (5.6351) acc 3.1250 (4.9219) lr 0.260000 +FPS@all 841.359, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:42 loss 5.8032 (5.7169) acc 0.0000 (4.5312) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.3269 (5.5991) acc 9.3750 (5.5469) lr 0.260000 +FPS@all 841.341, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.9521 (5.6746) acc 3.1250 (4.0625) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.2284 (5.5886) acc 6.2500 (4.6875) lr 0.260000 +FPS@all 841.393, TIME@all 0.304 +epoch: [2/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:28:41 loss 5.7224 (5.6761) acc 6.2500 (3.2812) lr 0.260000 +epoch: [2/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:28:27 loss 5.2207 (5.6229) acc 3.1250 (4.4531) lr 0.260000 +FPS@all 841.667, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:41 loss 6.0267 (5.6724) acc 3.1250 (2.3438) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:28:26 loss 5.7390 (5.6345) acc 9.3750 (3.6719) lr 0.260000 +FPS@all 841.543, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.7242 (5.7595) acc 3.1250 (2.5000) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.2590 (5.6375) acc 3.1250 (3.5156) lr 0.260000 +FPS@all 841.338, TIME@all 0.304 +epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:42 loss 6.2207 (5.7487) acc 0.0000 (3.4375) lr 0.260000 +epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:28:27 loss 5.6020 (5.6702) acc 6.2500 (4.3750) lr 0.260000 +FPS@all 841.475, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:27:57 loss 5.1104 (4.9978) acc 9.3750 (9.5312) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:27:59 loss 4.9897 (4.9629) acc 9.3750 (9.6875) lr 0.260000 +FPS@all 842.763, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:27:58 loss 5.0472 (4.9345) acc 15.6250 (10.0000) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:28:00 loss 4.9986 (4.9421) acc 12.5000 (10.7812) lr 0.260000 +FPS@all 842.666, TIME@all 0.304 +epoch: [3/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:27:58 loss 5.1691 (4.9396) acc 9.3750 (9.0625) lr 0.260000 +epoch: [3/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:27:59 loss 4.5438 (4.9401) acc 15.6250 (10.7031) lr 0.260000 +FPS@all 842.701, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:27:57 loss 5.0295 (4.9700) acc 9.3750 (7.0312) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:27:58 loss 4.8667 (5.0102) acc 9.3750 (8.2031) lr 0.260000 +FPS@all 842.872, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:58 loss 5.0562 (4.9847) acc 3.1250 (8.5938) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:28:00 loss 4.4445 (4.9519) acc 25.0000 (11.0156) lr 0.260000 +FPS@all 842.679, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:58 loss 4.8639 (5.0062) acc 6.2500 (10.3125) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:28:00 loss 4.6740 (4.9698) acc 9.3750 (10.9375) lr 0.260000 +FPS@all 842.677, TIME@all 0.304 +epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:57 loss 5.1656 (4.9917) acc 9.3750 (9.0625) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:27:59 loss 4.8903 (4.9856) acc 6.2500 (9.6094) lr 0.260000 +FPS@all 842.807, TIME@all 0.304 +epoch: [3/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:58 loss 5.4447 (5.0193) acc 0.0000 (7.5000) lr 0.260000 +epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:28:00 loss 4.6572 (5.0107) acc 15.6250 (9.4531) lr 0.260000 +FPS@all 843.073, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:27:49 loss 4.5347 (4.3350) acc 25.0000 (17.8125) lr 0.260000 +epoch: [4/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:27:49 loss 4.2485 (4.3576) acc 15.6250 (18.9844) lr 0.260000 +FPS@all 841.958, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:48 loss 4.5189 (4.3321) acc 21.8750 (18.4375) lr 0.260000 +epoch: [4/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.7004 (4.3285) acc 15.6250 (19.8438) lr 0.260000 +FPS@all 842.039, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.3967 (4.3375) acc 15.6250 (18.1250) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 3.9165 (4.3720) acc 31.2500 (19.2188) lr 0.260000 +FPS@all 841.983, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.3063 (4.3083) acc 15.6250 (20.7812) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.2638 (4.3564) acc 18.7500 (20.5469) lr 0.260000 +FPS@all 842.026, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.1575 (4.2676) acc 31.2500 (21.8750) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.7793 (4.3274) acc 21.8750 (19.6875) lr 0.260000 +FPS@all 841.969, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:27:48 loss 4.5746 (4.2994) acc 21.8750 (19.2188) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.3279 (4.4049) acc 25.0000 (17.8125) lr 0.260000 +FPS@all 842.127, TIME@all 0.304 +epoch: [4/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:27:47 loss 4.1815 (4.3585) acc 15.6250 (18.7500) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.4623 (4.3688) acc 25.0000 (19.7656) lr 0.260000 +FPS@all 842.311, TIME@all 0.304 +epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:27:47 loss 4.3963 (4.2453) acc 18.7500 (19.6875) lr 0.260000 +epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.4185 (4.3179) acc 18.7500 (19.3750) lr 0.260000 +FPS@all 842.184, TIME@all 0.304 +epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:27:36 loss 4.0796 (3.6327) acc 25.0000 (32.9688) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 4.4161 (3.7696) acc 12.5000 (31.0938) lr 0.260000 +FPS@all 842.687, TIME@all 0.304 +epoch: [5/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:27:35 loss 4.3063 (3.6727) acc 21.8750 (30.0000) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 3.9690 (3.7655) acc 37.5000 (31.0156) lr 0.260000 +FPS@all 842.738, TIME@all 0.304 +epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:27:36 loss 4.2187 (3.6356) acc 25.0000 (32.1875) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 3.6517 (3.7157) acc 37.5000 (32.5000) lr 0.260000 +FPS@all 842.732, TIME@all 0.304 +epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:36 loss 4.3091 (3.7495) acc 28.1250 (30.4688) lr 0.260000 +epoch: [5/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:30 loss 3.9241 (3.8058) acc 31.2500 (31.4062) lr 0.260000 +FPS@all 842.668, TIME@all 0.304 +epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:27:34 loss 3.1360 (3.6162) acc 37.5000 (34.6875) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.6961 (3.7158) acc 37.5000 (33.6719) lr 0.260000 +FPS@all 842.898, TIME@all 0.304 +epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:27:35 loss 3.7299 (3.6657) acc 25.0000 (32.1875) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:30 loss 3.3893 (3.7434) acc 50.0000 (33.1250) lr 0.260000 +FPS@all 842.690, TIME@all 0.304 +epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:34 loss 3.4656 (3.6112) acc 40.6250 (36.4062) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.8184 (3.7159) acc 37.5000 (35.7031) lr 0.260000 +FPS@all 842.850, TIME@all 0.304 +epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:35 loss 3.8033 (3.6672) acc 37.5000 (34.6875) lr 0.260000 +epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.7822 (3.7599) acc 21.8750 (32.2656) lr 0.260000 +FPS@all 843.088, TIME@all 0.304 +epoch: [6/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:27:17 loss 3.4712 (3.2225) acc 37.5000 (42.3438) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.2355 (3.3175) acc 46.8750 (42.0312) lr 0.260000 +FPS@all 844.135, TIME@all 0.303 +epoch: [6/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 1:27:17 loss 3.5757 (3.2662) acc 28.1250 (42.8125) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.8138 (3.3923) acc 34.3750 (39.9219) lr 0.260000 +FPS@all 844.166, TIME@all 0.303 +epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:16 loss 3.7874 (3.2680) acc 34.3750 (43.4375) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:06 loss 3.4127 (3.3299) acc 43.7500 (41.8750) lr 0.260000 +FPS@all 844.297, TIME@all 0.303 +epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:17 loss 3.5408 (3.2467) acc 40.6250 (43.5938) lr 0.260000 +epoch: [6/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:07 loss 3.4385 (3.3440) acc 37.5000 (41.6406) lr 0.260000 +FPS@all 844.129, TIME@all 0.303 +epoch: [6/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:27:17 loss 3.5238 (3.1541) acc 34.3750 (46.2500) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:07 loss 3.0333 (3.2738) acc 50.0000 (44.6094) lr 0.260000 +FPS@all 844.151, TIME@all 0.303 +epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:16 loss 3.7588 (3.2462) acc 31.2500 (44.8438) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:06 loss 3.6809 (3.3569) acc 28.1250 (42.5781) lr 0.260000 +FPS@all 844.330, TIME@all 0.303 +epoch: [6/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:27:16 loss 3.6366 (3.2437) acc 31.2500 (41.4062) lr 0.260000 +epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.4791 (3.3068) acc 28.1250 (40.6250) lr 0.260000 +FPS@all 844.182, TIME@all 0.303 +epoch: [6/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:27:18 loss 3.3052 (3.1660) acc 43.7500 (44.8438) lr 0.260000 +epoch: [6/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.6164 (3.2659) acc 40.6250 (43.0469) lr 0.260000 +FPS@all 844.436, TIME@all 0.303 +epoch: [7/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:50 loss 2.7394 (2.7289) acc 46.8750 (54.8438) lr 0.260000 +epoch: [7/350][40/50] time 0.306 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.2769 (2.9206) acc 43.7500 (50.7812) lr 0.260000 +FPS@all 844.501, TIME@all 0.303 +epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.6852 (2.7088) acc 53.1250 (58.7500) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.1574 (2.8643) acc 43.7500 (54.5312) lr 0.260000 +FPS@all 844.461, TIME@all 0.303 +epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 3.0215 (2.7730) acc 43.7500 (56.2500) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.2028 (2.9054) acc 43.7500 (53.6719) lr 0.260000 +FPS@all 844.508, TIME@all 0.303 +epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 1:26:48 loss 2.7774 (2.6933) acc 56.2500 (59.3750) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:45 loss 2.9467 (2.8475) acc 56.2500 (56.7969) lr 0.260000 +FPS@all 844.674, TIME@all 0.303 +epoch: [7/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 1:26:51 loss 3.0027 (2.7353) acc 50.0000 (54.5312) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.1123 (2.8863) acc 59.3750 (51.9531) lr 0.260000 +FPS@all 844.811, TIME@all 0.303 +epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.9843 (2.7426) acc 59.3750 (56.5625) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.3751 (2.8929) acc 53.1250 (53.9844) lr 0.260000 +FPS@all 844.469, TIME@all 0.303 +epoch: [7/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:48 loss 2.5921 (2.7193) acc 68.7500 (57.3438) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:45 loss 3.2694 (2.8827) acc 53.1250 (54.0625) lr 0.260000 +FPS@all 844.641, TIME@all 0.303 +epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.8563 (2.7895) acc 56.2500 (56.5625) lr 0.260000 +epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.7664 (2.9228) acc 37.5000 (52.8125) lr 0.260000 +FPS@all 844.499, TIME@all 0.303 +epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:27:00 loss 2.9792 (2.4935) acc 53.1250 (66.0938) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.8324 (2.6809) acc 56.2500 (60.6250) lr 0.260000 +FPS@all 842.667, TIME@all 0.304 +epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:26:59 loss 3.1148 (2.5098) acc 53.1250 (63.7500) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:26:45 loss 2.8392 (2.6727) acc 56.2500 (60.6250) lr 0.260000 +FPS@all 842.702, TIME@all 0.304 +epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:27:00 loss 3.0805 (2.5866) acc 50.0000 (60.3125) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:26:45 loss 2.7478 (2.7204) acc 59.3750 (56.6406) lr 0.260000 +FPS@all 842.592, TIME@all 0.304 +epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:26:59 loss 3.0695 (2.5173) acc 56.2500 (64.8438) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.6999 (2.6806) acc 59.3750 (60.8594) lr 0.260000 +FPS@all 842.623, TIME@all 0.304 +epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:26:58 loss 2.9053 (2.5056) acc 62.5000 (65.7812) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.5641 (2.6186) acc 59.3750 (62.1875) lr 0.260000 +FPS@all 842.831, TIME@all 0.304 +epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:26:59 loss 3.0868 (2.5277) acc 59.3750 (62.9688) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.5176 (2.6838) acc 59.3750 (58.2031) lr 0.260000 +FPS@all 842.627, TIME@all 0.304 +epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:26:57 loss 3.1584 (2.4755) acc 53.1250 (66.0938) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.3734 (2.6147) acc 68.7500 (61.3281) lr 0.260000 +FPS@all 843.071, TIME@all 0.304 +epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:26:58 loss 3.1165 (2.4870) acc 50.0000 (64.6875) lr 0.260000 +epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.7944 (2.6475) acc 53.1250 (59.5312) lr 0.260000 +FPS@all 842.780, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:26 loss 2.9404 (2.3622) acc 53.1250 (70.6250) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:26:24 loss 2.8924 (2.4852) acc 53.1250 (65.4688) lr 0.260000 +FPS@all 843.200, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:26:26 loss 3.0240 (2.3398) acc 56.2500 (70.4688) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.6576 (2.4789) acc 59.3750 (67.4219) lr 0.260000 +FPS@all 843.262, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:25 loss 3.2055 (2.3575) acc 53.1250 (69.0625) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:23 loss 2.7023 (2.5233) acc 59.3750 (65.0000) lr 0.260000 +FPS@all 843.411, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:25 loss 2.8743 (2.3435) acc 53.1250 (68.4375) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.8318 (2.4589) acc 59.3750 (65.4688) lr 0.260000 +FPS@all 843.267, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:26 loss 3.1199 (2.4655) acc 43.7500 (63.9062) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.9502 (2.5648) acc 56.2500 (62.9688) lr 0.260000 +FPS@all 843.218, TIME@all 0.304 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:25 loss 3.3551 (2.3044) acc 43.7500 (70.3125) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:23 loss 3.2092 (2.4686) acc 56.2500 (66.5625) lr 0.260000 +FPS@all 843.418, TIME@all 0.304 +epoch: [9/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:26:26 loss 3.4728 (2.3746) acc 50.0000 (66.4062) lr 0.260000 +epoch: [9/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:26:23 loss 3.1980 (2.4826) acc 43.7500 (64.7656) lr 0.260000 +FPS@all 843.549, TIME@all 0.303 +epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:26 loss 2.8887 (2.2948) acc 53.1250 (69.6875) lr 0.260000 +epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.7275 (2.4272) acc 53.1250 (66.4062) lr 0.260000 +FPS@all 843.220, TIME@all 0.304 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.6988 (2.0792) acc 53.1250 (75.3125) lr 0.260000 +epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.1807 (2.2598) acc 78.1250 (70.9375) lr 0.260000 +FPS@all 844.438, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.4798 (2.0968) acc 68.7500 (78.5938) lr 0.260000 +epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.4168 (2.2385) acc 59.3750 (73.0469) lr 0.260000 +FPS@all 844.346, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.3467 (2.0726) acc 78.1250 (77.1875) lr 0.260000 +epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:26:04 loss 2.4815 (2.2307) acc 65.6250 (72.4219) lr 0.260000 +FPS@all 844.407, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.3339 (2.0752) acc 78.1250 (76.8750) lr 0.260000 +epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:26:04 loss 2.0649 (2.2017) acc 78.1250 (73.9844) lr 0.260000 +FPS@all 844.389, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.4495 (2.0035) acc 71.8750 (80.6250) lr 0.260000 +epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.0464 (2.1636) acc 75.0000 (75.3125) lr 0.260000 +FPS@all 844.372, TIME@all 0.303 +epoch: [10/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.1854 (2.0741) acc 78.1250 (76.2500) lr 0.260000 +epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.2622 (2.2498) acc 71.8750 (72.5781) lr 0.260000 +FPS@all 844.738, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:05 loss 2.3676 (2.1259) acc 71.8750 (74.3750) lr 0.260000 +epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:26:03 loss 2.4101 (2.2517) acc 75.0000 (72.8906) lr 0.260000 +FPS@all 844.590, TIME@all 0.303 +epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:05 loss 2.5710 (2.1373) acc 62.5000 (75.6250) lr 0.260000 +epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:03 loss 2.5480 (2.2394) acc 68.7500 (74.1406) lr 0.260000 +FPS@all 844.531, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 1:25:58 loss 2.3500 (1.9106) acc 71.8750 (81.7188) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 1.9245 (2.1120) acc 78.1250 (75.9375) lr 0.260000 +FPS@all 843.800, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.4370 (1.8808) acc 56.2500 (82.3438) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.3339 (2.0881) acc 71.8750 (77.3438) lr 0.260000 +FPS@all 843.859, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:58 loss 2.0455 (1.8739) acc 78.1250 (82.3438) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.4180 (2.0505) acc 68.7500 (78.4375) lr 0.260000 +FPS@all 843.821, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.1906 (1.9447) acc 75.0000 (80.0000) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.2909 (2.0804) acc 75.0000 (76.9531) lr 0.260000 +FPS@all 843.824, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:25:56 loss 2.1193 (1.9145) acc 75.0000 (84.3750) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:50 loss 2.0402 (2.1367) acc 78.1250 (76.9531) lr 0.260000 +FPS@all 844.009, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 1.9884 (1.8950) acc 75.0000 (81.5625) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.4517 (2.1145) acc 65.6250 (76.6406) lr 0.260000 +FPS@all 843.813, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.3042 (1.9266) acc 68.7500 (80.6250) lr 0.260000 +epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:50 loss 2.2633 (2.1020) acc 68.7500 (76.0156) lr 0.260000 +FPS@all 843.954, TIME@all 0.303 +epoch: [11/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:25:57 loss 2.5468 (1.9712) acc 56.2500 (80.4688) lr 0.260000 +epoch: [11/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:50 loss 2.0838 (2.0805) acc 78.1250 (77.4219) lr 0.260000 +FPS@all 844.160, TIME@all 0.303 +epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:50 loss 2.3477 (1.9139) acc 68.7500 (82.6562) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.1936 (2.0725) acc 68.7500 (77.9688) lr 0.260000 +FPS@all 842.655, TIME@all 0.304 +epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:50 loss 2.1892 (1.9188) acc 84.3750 (82.0312) lr 0.260000 +epoch: [12/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:25:46 loss 2.3637 (2.0791) acc 68.7500 (77.4219) lr 0.260000 +FPS@all 842.589, TIME@all 0.304 +epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:50 loss 2.1258 (1.9314) acc 84.3750 (81.2500) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.3672 (2.0763) acc 65.6250 (76.4062) lr 0.260000 +FPS@all 842.590, TIME@all 0.304 +epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.5196 (1.9502) acc 59.3750 (80.1562) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:45 loss 2.3533 (2.0755) acc 59.3750 (76.9531) lr 0.260000 +FPS@all 842.808, TIME@all 0.304 +epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:25:49 loss 2.0882 (1.9053) acc 78.1250 (83.5938) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.2517 (2.0403) acc 71.8750 (79.2969) lr 0.260000 +FPS@all 842.615, TIME@all 0.304 +epoch: [12/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:25:50 loss 2.1714 (1.8672) acc 78.1250 (83.4375) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:46 loss 2.2614 (2.1188) acc 71.8750 (76.0938) lr 0.260000 +FPS@all 842.606, TIME@all 0.304 +epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.4305 (1.9507) acc 68.7500 (80.9375) lr 0.260000 +epoch: [12/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:25:45 loss 2.3816 (2.1251) acc 68.7500 (76.6406) lr 0.260000 +FPS@all 842.759, TIME@all 0.304 +epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.0246 (1.8819) acc 78.1250 (84.2188) lr 0.260000 +epoch: [12/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:25:45 loss 1.9955 (2.0819) acc 75.0000 (78.8281) lr 0.260000 +FPS@all 842.996, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:28 loss 2.1344 (1.8721) acc 78.1250 (83.9062) lr 0.260000 +epoch: [13/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0766 (1.9707) acc 75.0000 (79.7656) lr 0.260000 +FPS@all 842.959, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:29 loss 2.2122 (1.8762) acc 62.5000 (82.5000) lr 0.260000 +epoch: [13/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0595 (1.9928) acc 75.0000 (80.4688) lr 0.260000 +FPS@all 842.901, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 1:25:29 loss 2.2509 (1.8982) acc 71.8750 (83.7500) lr 0.260000 +epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.2267 (1.9728) acc 75.0000 (80.7031) lr 0.260000 +FPS@all 842.874, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:29 loss 2.2561 (1.9078) acc 68.7500 (82.3438) lr 0.260000 +epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:25:26 loss 2.1083 (1.9869) acc 71.8750 (80.0781) lr 0.260000 +FPS@all 842.884, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:25:27 loss 2.0696 (1.8892) acc 75.0000 (82.9688) lr 0.260000 +epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:25:25 loss 1.9480 (1.9883) acc 84.3750 (79.1406) lr 0.260000 +FPS@all 843.092, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:29 loss 2.2525 (1.8713) acc 84.3750 (85.3125) lr 0.260000 +epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:26 loss 1.9996 (1.9672) acc 78.1250 (81.4062) lr 0.260000 +FPS@all 842.893, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:28 loss 2.0159 (1.8965) acc 81.2500 (84.2188) lr 0.260000 +epoch: [13/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:25:25 loss 1.9084 (1.9571) acc 84.3750 (82.0312) lr 0.260000 +FPS@all 843.018, TIME@all 0.304 +epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:27 loss 2.2537 (1.8796) acc 71.8750 (82.3438) lr 0.260000 +epoch: [13/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0106 (1.9679) acc 81.2500 (80.2344) lr 0.260000 +FPS@all 843.231, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.011) eta 1:25:39 loss 2.0438 (1.7822) acc 81.2500 (86.0938) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.0686 (1.8955) acc 75.0000 (82.4219) lr 0.260000 +FPS@all 842.934, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.001 (0.011) eta 1:25:38 loss 2.1837 (1.7730) acc 75.0000 (87.8125) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.1743 (1.8804) acc 78.1250 (84.0625) lr 0.260000 +FPS@all 842.978, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.011) eta 1:25:38 loss 2.2646 (1.7770) acc 71.8750 (86.2500) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.2471 (1.9031) acc 75.0000 (82.9688) lr 0.260000 +FPS@all 843.044, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3168 (1.7839) acc 75.0000 (84.3750) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.3512 (1.8856) acc 62.5000 (81.4062) lr 0.260000 +FPS@all 842.962, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3447 (1.8452) acc 71.8750 (82.6562) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.0083 (1.9111) acc 78.1250 (81.9531) lr 0.260000 +FPS@all 842.955, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:25:37 loss 2.2072 (1.7927) acc 71.8750 (87.3438) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:13 loss 1.8632 (1.8792) acc 81.2500 (83.6719) lr 0.260000 +FPS@all 843.169, TIME@all 0.304 +epoch: [14/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:25:38 loss 1.9521 (1.7738) acc 87.5000 (86.0938) lr 0.260000 +epoch: [14/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.2463 (1.8876) acc 71.8750 (82.2656) lr 0.260000 +FPS@all 843.332, TIME@all 0.304 +epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3742 (1.7801) acc 68.7500 (85.9375) lr 0.260000 +epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:13 loss 1.9280 (1.8951) acc 75.0000 (82.0312) lr 0.260000 +FPS@all 843.106, TIME@all 0.304 +epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.011) eta 1:24:43 loss 1.9269 (1.7250) acc 90.6250 (89.2188) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 2.0384 (1.8059) acc 75.0000 (85.5469) lr 0.260000 +FPS@all 844.099, TIME@all 0.303 +epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.9626 (1.7266) acc 78.1250 (89.0625) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.6979 (1.8216) acc 93.7500 (85.2344) lr 0.260000 +FPS@all 844.128, TIME@all 0.303 +epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:24:42 loss 1.6664 (1.7338) acc 87.5000 (87.0312) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 2.1876 (1.8102) acc 78.1250 (84.2969) lr 0.260000 +FPS@all 844.144, TIME@all 0.303 +epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.9600 (1.7508) acc 78.1250 (87.3438) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 1.8862 (1.8121) acc 81.2500 (85.0000) lr 0.260000 +FPS@all 844.110, TIME@all 0.303 +epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:24:42 loss 1.6719 (1.7434) acc 90.6250 (87.8125) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:24:46 loss 1.7332 (1.8116) acc 93.7500 (85.2344) lr 0.260000 +FPS@all 844.321, TIME@all 0.303 +epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:24:42 loss 1.8860 (1.7333) acc 84.3750 (87.6562) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.5067 (1.8059) acc 96.8750 (85.5469) lr 0.260000 +FPS@all 844.269, TIME@all 0.303 +epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.7712 (1.7148) acc 87.5000 (87.5000) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 1.6564 (1.7955) acc 93.7500 (84.8438) lr 0.260000 +FPS@all 844.099, TIME@all 0.303 +epoch: [15/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 1:24:42 loss 1.7334 (1.7192) acc 87.5000 (89.6875) lr 0.260000 +epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.7172 (1.8116) acc 87.5000 (85.4688) lr 0.260000 +FPS@all 844.484, TIME@all 0.303 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.8232 (1.6718) acc 84.3750 (89.6875) lr 0.260000 +epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.7307 (1.7648) acc 87.5000 (85.7031) lr 0.260000 +FPS@all 842.654, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 2.2704 (1.7408) acc 68.7500 (85.9375) lr 0.260000 +epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.8284 (1.8023) acc 84.3750 (84.2188) lr 0.260000 +FPS@all 842.709, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.9320 (1.6923) acc 87.5000 (89.3750) lr 0.260000 +epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 2.0041 (1.7521) acc 78.1250 (87.5781) lr 0.260000 +FPS@all 842.681, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:53 loss 1.9430 (1.6440) acc 90.6250 (91.7188) lr 0.260000 +epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.8862 (1.7211) acc 78.1250 (88.5156) lr 0.260000 +FPS@all 842.678, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:52 loss 2.0422 (1.7293) acc 81.2500 (88.7500) lr 0.260000 +epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:42 loss 1.8093 (1.7598) acc 87.5000 (87.1875) lr 0.260000 +FPS@all 842.837, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:52 loss 1.9518 (1.6579) acc 78.1250 (88.5938) lr 0.260000 +epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:42 loss 1.8418 (1.7446) acc 84.3750 (86.0156) lr 0.260000 +FPS@all 842.879, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.7201 (1.6666) acc 84.3750 (87.0312) lr 0.260000 +epoch: [16/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:24:43 loss 1.7872 (1.7213) acc 84.3750 (86.1719) lr 0.260000 +FPS@all 842.710, TIME@all 0.304 +epoch: [16/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:24:52 loss 1.6927 (1.6364) acc 87.5000 (88.2812) lr 0.260000 +epoch: [16/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:24:42 loss 1.9825 (1.7273) acc 68.7500 (85.7812) lr 0.260000 +FPS@all 843.087, TIME@all 0.304 +epoch: [17/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6068 (1.6308) acc 90.6250 (91.0938) lr 0.260000 +epoch: [17/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.9334 (1.6947) acc 75.0000 (88.3594) lr 0.260000 +FPS@all 842.256, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.8528 (1.6177) acc 87.5000 (90.0000) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:35 loss 2.1174 (1.7142) acc 71.8750 (87.3438) lr 0.260000 +FPS@all 842.178, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:37 loss 1.7029 (1.6356) acc 87.5000 (90.6250) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:24:33 loss 2.0299 (1.7207) acc 84.3750 (87.9688) lr 0.260000 +FPS@all 842.390, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:38 loss 1.7138 (1.6227) acc 84.3750 (91.0938) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.9223 (1.7092) acc 75.0000 (87.5000) lr 0.260000 +FPS@all 842.187, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6418 (1.6722) acc 90.6250 (89.2188) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.8367 (1.7258) acc 84.3750 (87.5000) lr 0.260000 +FPS@all 842.534, TIME@all 0.304 +epoch: [17/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6256 (1.6194) acc 87.5000 (90.9375) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.6249 (1.7132) acc 84.3750 (88.3594) lr 0.260000 +FPS@all 842.212, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:37 loss 1.8583 (1.6009) acc 81.2500 (91.0938) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:24:34 loss 1.9272 (1.6970) acc 87.5000 (88.2812) lr 0.260000 +FPS@all 842.352, TIME@all 0.304 +epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6337 (1.6059) acc 87.5000 (90.0000) lr 0.260000 +epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.7891 (1.6995) acc 81.2500 (87.5781) lr 0.260000 +FPS@all 842.202, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:16 loss 1.6517 (1.6342) acc 87.5000 (89.8438) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:24:13 loss 1.5770 (1.6491) acc 84.3750 (89.2188) lr 0.260000 +FPS@all 841.871, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.8928 (1.6250) acc 84.3750 (90.3125) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:13 loss 1.6660 (1.6828) acc 93.7500 (88.5938) lr 0.260000 +FPS@all 841.909, TIME@all 0.304 +epoch: [18/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.8458 (1.6233) acc 84.3750 (90.1562) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.5797 (1.6621) acc 90.6250 (88.3594) lr 0.260000 +FPS@all 841.924, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.6356 (1.5906) acc 93.7500 (92.5000) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.4996 (1.6321) acc 96.8750 (90.3906) lr 0.260000 +FPS@all 841.863, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.7957 (1.6170) acc 84.3750 (89.0625) lr 0.260000 +epoch: [18/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.5924 (1.6327) acc 90.6250 (89.6875) lr 0.260000 +FPS@all 841.871, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:24:15 loss 1.6906 (1.6183) acc 90.6250 (89.3750) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:12 loss 1.5967 (1.6827) acc 87.5000 (87.2656) lr 0.260000 +FPS@all 842.051, TIME@all 0.304 +epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:15 loss 1.8373 (1.6506) acc 87.5000 (90.1562) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:12 loss 1.6518 (1.6625) acc 87.5000 (90.0000) lr 0.260000 +FPS@all 841.997, TIME@all 0.304 +epoch: [18/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:24:16 loss 2.0002 (1.6281) acc 84.3750 (90.1562) lr 0.260000 +epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:13 loss 1.6149 (1.6387) acc 90.6250 (89.9219) lr 0.260000 +FPS@all 842.202, TIME@all 0.304 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:59 loss 1.6540 (1.5709) acc 90.6250 (90.7812) lr 0.260000 +epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:23:51 loss 1.6243 (1.6345) acc 84.3750 (88.4375) lr 0.260000 +FPS@all 844.138, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:00 loss 1.6576 (1.5697) acc 84.3750 (90.3125) lr 0.260000 +epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.6060 (1.6423) acc 90.6250 (89.0625) lr 0.260000 +FPS@all 844.159, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:00 loss 1.8525 (1.5766) acc 81.2500 (90.3125) lr 0.260000 +epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:23:51 loss 1.4212 (1.6297) acc 96.8750 (89.0625) lr 0.260000 +FPS@all 844.170, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:59 loss 2.1074 (1.6043) acc 75.0000 (90.6250) lr 0.260000 +epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.5387 (1.6736) acc 90.6250 (88.2031) lr 0.260000 +FPS@all 844.314, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:23:58 loss 1.8209 (1.5300) acc 78.1250 (92.9688) lr 0.260000 +epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.4963 (1.5918) acc 93.7500 (91.8750) lr 0.260000 +FPS@all 844.358, TIME@all 0.303 +epoch: [19/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:23:57 loss 1.8268 (1.5641) acc 84.3750 (92.1875) lr 0.260000 +epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.6087 (1.6355) acc 87.5000 (89.0625) lr 0.260000 +FPS@all 844.445, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:00 loss 1.6196 (1.5919) acc 87.5000 (90.7812) lr 0.260000 +epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.5862 (1.6471) acc 90.6250 (88.9844) lr 0.260000 +FPS@all 844.171, TIME@all 0.303 +epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:59 loss 1.7431 (1.5879) acc 78.1250 (89.0625) lr 0.260000 +epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.8701 (1.6353) acc 81.2500 (88.7500) lr 0.260000 +FPS@all 844.173, TIME@all 0.303 +epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6368 (1.5870) acc 90.6250 (90.6250) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.4452 (1.5943) acc 100.0000 (90.5469) lr 0.260000 +FPS@all 843.509, TIME@all 0.303 +epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:49 loss 1.6363 (1.5571) acc 90.6250 (92.3438) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.6918 (1.6112) acc 90.6250 (90.0781) lr 0.260000 +FPS@all 843.468, TIME@all 0.304 +epoch: [20/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:23:49 loss 1.5494 (1.5056) acc 93.7500 (92.8125) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.7375 (1.5875) acc 84.3750 (90.1562) lr 0.260000 +FPS@all 843.395, TIME@all 0.304 +epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:49 loss 1.6179 (1.5562) acc 90.6250 (91.8750) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.4542 (1.6301) acc 96.8750 (89.7656) lr 0.260000 +FPS@all 843.438, TIME@all 0.304 +epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:23:48 loss 1.7041 (1.5407) acc 87.5000 (90.7812) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:36 loss 1.7553 (1.5971) acc 87.5000 (89.6875) lr 0.260000 +FPS@all 843.632, TIME@all 0.303 +epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6396 (1.5228) acc 93.7500 (93.1250) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.5644 (1.5941) acc 90.6250 (91.0938) lr 0.260000 +FPS@all 843.447, TIME@all 0.304 +epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6168 (1.5395) acc 87.5000 (92.3438) lr 0.260000 +epoch: [20/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.5754 (1.5801) acc 96.8750 (90.9375) lr 0.260000 +FPS@all 843.756, TIME@all 0.303 +epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.7928 (1.5319) acc 84.3750 (90.9375) lr 0.260000 +epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:36 loss 1.6874 (1.6130) acc 90.6250 (89.2188) lr 0.260000 +FPS@all 843.565, TIME@all 0.303 +epoch: [21/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:32 loss 1.6609 (1.4867) acc 87.5000 (94.3750) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.5925 (1.5507) acc 90.6250 (92.0312) lr 0.260000 +FPS@all 842.783, TIME@all 0.304 +epoch: [21/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:33 loss 1.6264 (1.5211) acc 90.6250 (92.6562) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.7088 (1.5811) acc 87.5000 (91.0156) lr 0.260000 +FPS@all 842.739, TIME@all 0.304 +epoch: [21/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:32 loss 1.5246 (1.5038) acc 93.7500 (92.8125) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:23 loss 1.6952 (1.5606) acc 96.8750 (91.8750) lr 0.260000 +FPS@all 842.785, TIME@all 0.304 +epoch: [21/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.4524 (1.4895) acc 96.8750 (93.4375) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:24 loss 1.6472 (1.5544) acc 90.6250 (91.3281) lr 0.260000 +FPS@all 842.775, TIME@all 0.304 +epoch: [21/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:23:31 loss 1.6183 (1.5179) acc 87.5000 (92.5000) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:22 loss 1.7229 (1.5657) acc 81.2500 (91.5625) lr 0.260000 +FPS@all 842.969, TIME@all 0.304 +epoch: [21/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.3999 (1.5017) acc 96.8750 (92.9688) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:24 loss 1.6378 (1.5621) acc 87.5000 (91.5625) lr 0.260000 +FPS@all 842.771, TIME@all 0.304 +epoch: [21/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.6774 (1.5058) acc 87.5000 (93.2812) lr 0.260000 +epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:23 loss 1.8060 (1.5426) acc 84.3750 (92.4219) lr 0.260000 +FPS@all 842.925, TIME@all 0.304 +epoch: [21/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:33 loss 1.5637 (1.5081) acc 90.6250 (93.1250) lr 0.260000 +epoch: [21/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.6745 (1.5669) acc 84.3750 (91.6406) lr 0.260000 +FPS@all 843.104, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4547 (1.4388) acc 90.6250 (94.3750) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5501 (1.5267) acc 90.6250 (92.5781) lr 0.260000 +FPS@all 841.927, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4106 (1.4491) acc 96.8750 (94.5312) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5252 (1.5035) acc 96.8750 (93.2812) lr 0.260000 +FPS@all 841.989, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4990 (1.4538) acc 93.7500 (94.8438) lr 0.260000 +epoch: [22/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:23:15 loss 1.4044 (1.5409) acc 96.8750 (92.5000) lr 0.260000 +FPS@all 841.962, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.5454 (1.4536) acc 90.6250 (94.3750) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5478 (1.5200) acc 93.7500 (92.9688) lr 0.260000 +FPS@all 841.960, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:23:24 loss 1.5106 (1.4572) acc 90.6250 (94.2188) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:14 loss 1.5968 (1.5300) acc 90.6250 (92.5000) lr 0.260000 +FPS@all 842.155, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:23:24 loss 1.6081 (1.5177) acc 93.7500 (93.2812) lr 0.260000 +epoch: [22/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:15 loss 1.7207 (1.5342) acc 84.3750 (93.0469) lr 0.260000 +FPS@all 842.318, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.6046 (1.4892) acc 90.6250 (93.5938) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:16 loss 1.7763 (1.5437) acc 84.3750 (91.4844) lr 0.260000 +FPS@all 841.945, TIME@all 0.304 +epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:23:24 loss 1.4799 (1.4689) acc 90.6250 (95.3125) lr 0.260000 +epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:15 loss 1.5760 (1.5538) acc 90.6250 (91.8750) lr 0.260000 +FPS@all 842.080, TIME@all 0.304 +epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 1:22:49 loss 1.6826 (1.4862) acc 93.7500 (92.9688) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:47 loss 1.5364 (1.5273) acc 87.5000 (92.1094) lr 0.260000 +FPS@all 844.241, TIME@all 0.303 +epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 1:22:48 loss 1.3989 (1.4764) acc 96.8750 (92.9688) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:46 loss 1.4861 (1.5280) acc 96.8750 (91.7969) lr 0.260000 +FPS@all 844.320, TIME@all 0.303 +epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:22:49 loss 1.5566 (1.4916) acc 90.6250 (92.9688) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5078 (1.5196) acc 93.7500 (92.0312) lr 0.260000 +FPS@all 844.275, TIME@all 0.303 +epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.5836 (1.4664) acc 93.7500 (94.2188) lr 0.260000 +epoch: [23/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5335 (1.5126) acc 90.6250 (92.6562) lr 0.260000 +FPS@all 844.281, TIME@all 0.303 +epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 1:22:47 loss 1.5609 (1.4499) acc 90.6250 (95.3125) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:45 loss 1.5147 (1.5287) acc 90.6250 (92.7344) lr 0.260000 +FPS@all 844.469, TIME@all 0.303 +epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.4855 (1.5202) acc 90.6250 (92.5000) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.3572 (1.5464) acc 100.0000 (92.1875) lr 0.260000 +FPS@all 844.280, TIME@all 0.303 +epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.4994 (1.4266) acc 90.6250 (95.1562) lr 0.260000 +epoch: [23/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.6486 (1.5329) acc 87.5000 (92.1875) lr 0.260000 +FPS@all 844.569, TIME@all 0.303 +epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.5796 (1.4849) acc 93.7500 (94.3750) lr 0.260000 +epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5325 (1.5279) acc 87.5000 (92.2656) lr 0.260000 +FPS@all 844.394, TIME@all 0.303 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 1:22:47 loss 1.4438 (1.4761) acc 93.7500 (93.2812) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:42 loss 1.6138 (1.5045) acc 90.6250 (92.1094) lr 0.260000 +FPS@all 843.258, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.5758 (1.4738) acc 90.6250 (93.5938) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.4847 (1.5046) acc 93.7500 (92.7344) lr 0.260000 +FPS@all 843.292, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 1:22:46 loss 1.4917 (1.5130) acc 93.7500 (92.8125) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:42 loss 1.5912 (1.5180) acc 93.7500 (92.8125) lr 0.260000 +FPS@all 843.313, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.4771 (1.4776) acc 90.6250 (93.4375) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.7465 (1.5324) acc 81.2500 (91.7969) lr 0.260000 +FPS@all 843.288, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 1:22:46 loss 1.5804 (1.4843) acc 90.6250 (93.2812) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.4862 (1.4840) acc 93.7500 (93.3594) lr 0.260000 +FPS@all 843.424, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:47 loss 1.6834 (1.4677) acc 93.7500 (94.3750) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.4685 (1.5081) acc 93.7500 (93.0469) lr 0.260000 +FPS@all 843.268, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.001 (0.014) eta 1:22:45 loss 1.4580 (1.4679) acc 93.7500 (94.0625) lr 0.260000 +epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.7204 (1.5035) acc 84.3750 (92.8125) lr 0.260000 +FPS@all 843.483, TIME@all 0.304 +epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.5345 (1.4554) acc 87.5000 (94.3750) lr 0.260000 +epoch: [24/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.3837 (1.4919) acc 93.7500 (93.2031) lr 0.260000 +FPS@all 843.668, TIME@all 0.303 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5613 (1.4574) acc 93.7500 (95.1562) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.6645 (1.5084) acc 84.3750 (93.0469) lr 0.260000 +FPS@all 840.948, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:22:42 loss 1.6904 (1.5296) acc 90.6250 (92.5000) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:22:38 loss 1.5665 (1.5438) acc 93.7500 (91.4844) lr 0.260000 +FPS@all 840.910, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:22:41 loss 1.5668 (1.4762) acc 96.8750 (94.0625) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:22:37 loss 1.5616 (1.5334) acc 90.6250 (92.5000) lr 0.260000 +FPS@all 840.974, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5881 (1.4786) acc 87.5000 (93.5938) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:38 loss 1.4845 (1.4953) acc 87.5000 (92.7344) lr 0.260000 +FPS@all 840.924, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5575 (1.4775) acc 90.6250 (93.4375) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.5321 (1.5220) acc 90.6250 (92.4219) lr 0.260000 +FPS@all 841.077, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5106 (1.5379) acc 100.0000 (92.1875) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:38 loss 1.5961 (1.5574) acc 87.5000 (91.2500) lr 0.260000 +FPS@all 840.921, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:42 loss 1.5237 (1.4779) acc 93.7500 (94.8438) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.5036 (1.5181) acc 93.7500 (93.4375) lr 0.260000 +FPS@all 841.256, TIME@all 0.304 +epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:40 loss 1.5909 (1.5067) acc 87.5000 (92.8125) lr 0.260000 +epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:36 loss 1.6199 (1.5342) acc 90.6250 (91.9531) lr 0.260000 +FPS@all 841.136, TIME@all 0.304 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:22:15 loss 1.4695 (1.4284) acc 90.6250 (95.1562) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:02 loss 1.5281 (1.4940) acc 93.7500 (93.2031) lr 0.260000 +FPS@all 844.639, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4233 (1.4277) acc 100.0000 (95.3125) lr 0.260000 +epoch: [26/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.6222 (1.4897) acc 87.5000 (93.5156) lr 0.260000 +FPS@all 844.690, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:22:14 loss 1.4912 (1.4271) acc 90.6250 (95.0000) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:02 loss 1.4413 (1.4866) acc 96.8750 (92.6562) lr 0.260000 +FPS@all 844.685, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4756 (1.4871) acc 90.6250 (93.2812) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.5794 (1.5400) acc 87.5000 (91.2500) lr 0.260000 +FPS@all 844.648, TIME@all 0.303 +epoch: [26/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4982 (1.4342) acc 93.7500 (95.3125) lr 0.260000 +epoch: [26/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:22:00 loss 1.3261 (1.4772) acc 96.8750 (93.7500) lr 0.260000 +FPS@all 845.017, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4850 (1.4275) acc 96.8750 (94.8438) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:01 loss 1.4930 (1.4946) acc 93.7500 (92.6562) lr 0.260000 +FPS@all 844.810, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:14 loss 1.6054 (1.4459) acc 90.6250 (94.3750) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:01 loss 1.6510 (1.5111) acc 90.6250 (92.5781) lr 0.260000 +FPS@all 844.856, TIME@all 0.303 +epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.3985 (1.4197) acc 90.6250 (93.2812) lr 0.260000 +epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.4797 (1.4796) acc 96.8750 (92.3438) lr 0.260000 +FPS@all 844.669, TIME@all 0.303 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.5097 (1.4394) acc 90.6250 (94.5312) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:22:04 loss 1.4779 (1.4942) acc 93.7500 (92.9688) lr 0.260000 +FPS@all 842.440, TIME@all 0.304 +epoch: [27/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.4414 (1.4462) acc 96.8750 (93.9062) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.4625 (1.5148) acc 93.7500 (91.1719) lr 0.260000 +FPS@all 842.438, TIME@all 0.304 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:22:12 loss 1.5433 (1.4570) acc 90.6250 (94.5312) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:22:03 loss 1.5172 (1.4869) acc 90.6250 (93.8281) lr 0.260000 +FPS@all 842.525, TIME@all 0.304 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.4825 (1.4180) acc 93.7500 (95.6250) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.4554 (1.4910) acc 90.6250 (93.1250) lr 0.260000 +FPS@all 842.440, TIME@all 0.304 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.5227 (1.4483) acc 93.7500 (93.2812) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.5613 (1.4768) acc 90.6250 (93.4375) lr 0.260000 +FPS@all 842.450, TIME@all 0.304 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:11 loss 1.4760 (1.4635) acc 90.6250 (93.5938) lr 0.260000 +epoch: [27/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.7285 (1.5040) acc 84.3750 (92.1094) lr 0.260000 +FPS@all 842.816, TIME@all 0.304 +epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:22:11 loss 1.4941 (1.4683) acc 96.8750 (95.0000) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:03 loss 1.6245 (1.5023) acc 90.6250 (93.1250) lr 0.260000 +FPS@all 842.654, TIME@all 0.304 +epoch: [27/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:22:11 loss 1.6080 (1.4627) acc 87.5000 (94.5312) lr 0.260000 +epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:03 loss 1.4639 (1.5207) acc 96.8750 (92.1875) lr 0.260000 +FPS@all 842.614, TIME@all 0.304 +epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.5383 (1.4260) acc 87.5000 (94.6875) lr 0.260000 +epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.4964 (1.4809) acc 90.6250 (93.6719) lr 0.260000 +FPS@all 842.596, TIME@all 0.304 +epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:21:44 loss 1.3346 (1.4317) acc 96.8750 (95.1562) lr 0.260000 +epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3215 (1.4743) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 842.497, TIME@all 0.304 +epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:21:43 loss 1.3860 (1.4337) acc 96.8750 (94.5312) lr 0.260000 +epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.4900 (1.4702) acc 90.6250 (94.1406) lr 0.260000 +FPS@all 842.566, TIME@all 0.304 +epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:21:44 loss 1.3578 (1.4242) acc 96.8750 (94.6875) lr 0.260000 +epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3231 (1.4488) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 842.525, TIME@all 0.304 +epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.3018 (1.4068) acc 96.8750 (95.4688) lr 0.260000 +epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3695 (1.4556) acc 93.7500 (93.5156) lr 0.260000 +FPS@all 842.510, TIME@all 0.304 +epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:21:43 loss 1.2834 (1.4353) acc 96.8750 (94.5312) lr 0.260000 +epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:21:41 loss 1.4604 (1.4591) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 842.648, TIME@all 0.304 +epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:21:43 loss 1.3176 (1.4344) acc 100.0000 (95.1562) lr 0.260000 +epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:21:41 loss 1.3464 (1.4445) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 842.705, TIME@all 0.304 +epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.4319 (1.4594) acc 96.8750 (94.8438) lr 0.260000 +epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3979 (1.4787) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 842.826, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 1:21:46 loss 1.4040 (1.4694) acc 93.7500 (94.6875) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.4826 (1.4985) acc 90.6250 (93.4375) lr 0.260000 +FPS@all 841.308, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:46 loss 1.5328 (1.4742) acc 93.7500 (93.9062) lr 0.260000 +epoch: [29/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.4068 (1.4879) acc 93.7500 (93.5938) lr 0.260000 +FPS@all 841.322, TIME@all 0.304 +epoch: [29/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:21:45 loss 1.3712 (1.4458) acc 96.8750 (94.6875) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.5679 (1.5027) acc 93.7500 (92.5781) lr 0.260000 +FPS@all 841.347, TIME@all 0.304 +epoch: [29/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.2658 (1.4347) acc 100.0000 (95.1562) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:36 loss 1.4188 (1.4724) acc 96.8750 (93.5156) lr 0.260000 +FPS@all 841.316, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.4594 (1.4482) acc 93.7500 (94.2188) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:36 loss 1.5041 (1.4841) acc 93.7500 (93.4375) lr 0.260000 +FPS@all 841.332, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:44 loss 1.5366 (1.4618) acc 93.7500 (94.2188) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.3773 (1.4778) acc 93.7500 (93.5156) lr 0.260000 +FPS@all 841.709, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.6286 (1.4389) acc 90.6250 (94.8438) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.3795 (1.4809) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 841.460, TIME@all 0.304 +epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 1:21:44 loss 1.4948 (1.4753) acc 87.5000 (93.9062) lr 0.260000 +epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.4476 (1.4994) acc 96.8750 (92.8125) lr 0.260000 +FPS@all 841.514, TIME@all 0.304 +epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 1:21:13 loss 1.4448 (1.4722) acc 96.8750 (93.2812) lr 0.260000 +epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:21:10 loss 1.4200 (1.5036) acc 96.8750 (92.1875) lr 0.260000 +FPS@all 843.450, TIME@all 0.304 +epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 1:21:11 loss 1.5068 (1.4800) acc 93.7500 (93.9062) lr 0.260000 +epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:21:09 loss 1.4946 (1.5116) acc 93.7500 (92.7344) lr 0.260000 +FPS@all 843.592, TIME@all 0.303 +epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.6495 (1.4886) acc 90.6250 (94.8438) lr 0.260000 +epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:10 loss 1.4396 (1.5168) acc 93.7500 (93.2812) lr 0.260000 +FPS@all 843.478, TIME@all 0.304 +epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.7469 (1.4573) acc 90.6250 (94.6875) lr 0.260000 +epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 1:21:10 loss 1.4834 (1.4990) acc 93.7500 (93.3594) lr 0.260000 +FPS@all 843.474, TIME@all 0.304 +epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:12 loss 1.4647 (1.4640) acc 90.6250 (93.7500) lr 0.260000 +epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.4799 (1.5009) acc 100.0000 (92.5000) lr 0.260000 +FPS@all 843.658, TIME@all 0.303 +epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.6040 (1.4694) acc 87.5000 (93.7500) lr 0.260000 +epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.5660 (1.4948) acc 90.6250 (92.8125) lr 0.260000 +FPS@all 843.614, TIME@all 0.303 +epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:21:12 loss 1.4890 (1.4655) acc 90.6250 (94.2188) lr 0.260000 +epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.3320 (1.5026) acc 100.0000 (92.6562) lr 0.260000 +FPS@all 843.809, TIME@all 0.303 +epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.5546 (1.4493) acc 90.6250 (93.4375) lr 0.260000 +epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:21:09 loss 1.4316 (1.5283) acc 96.8750 (91.8750) lr 0.260000 +FPS@all 843.511, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:21:04 loss 1.4819 (1.3867) acc 96.8750 (97.0312) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:20:54 loss 1.4693 (1.4344) acc 90.6250 (94.7656) lr 0.260000 +FPS@all 843.471, TIME@all 0.304 +epoch: [31/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.5075 (1.4360) acc 93.7500 (94.6875) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:20:54 loss 1.3717 (1.4532) acc 100.0000 (94.1406) lr 0.260000 +FPS@all 843.540, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:21:04 loss 1.6758 (1.4415) acc 90.6250 (95.4688) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:20:54 loss 1.2680 (1.4672) acc 100.0000 (94.0625) lr 0.260000 +FPS@all 843.515, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.4401 (1.4132) acc 87.5000 (94.2188) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.4437 (1.4692) acc 96.8750 (93.2031) lr 0.260000 +FPS@all 843.633, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:05 loss 1.4271 (1.4243) acc 93.7500 (93.4375) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:20:55 loss 1.4306 (1.4675) acc 96.8750 (92.8906) lr 0.260000 +FPS@all 843.463, TIME@all 0.304 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:03 loss 1.5094 (1.4319) acc 93.7500 (94.8438) lr 0.260000 +epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.4957 (1.4566) acc 93.7500 (94.0625) lr 0.260000 +FPS@all 843.669, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 1:21:05 loss 1.4201 (1.4312) acc 93.7500 (95.3125) lr 0.260000 +epoch: [31/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.3450 (1.4602) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 843.810, TIME@all 0.303 +epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.4074 (1.4278) acc 96.8750 (94.8438) lr 0.260000 +epoch: [31/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:20:54 loss 1.4347 (1.4551) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 843.464, TIME@all 0.304 +epoch: [32/350][20/50] time 0.308 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.3327 (1.3897) acc 100.0000 (95.7812) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.6014 (1.4235) acc 87.5000 (94.5312) lr 0.260000 +FPS@all 843.763, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.011) eta 1:20:33 loss 1.5115 (1.4345) acc 90.6250 (93.9062) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.3753 (1.4714) acc 100.0000 (92.9688) lr 0.260000 +FPS@all 843.790, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.011) eta 1:20:33 loss 1.4833 (1.3992) acc 96.8750 (95.3125) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.4745 (1.4309) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 843.718, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.001 (0.012) eta 1:20:33 loss 1.8234 (1.4170) acc 90.6250 (95.7812) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:32 loss 1.3378 (1.4278) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 843.788, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.6402 (1.4190) acc 90.6250 (94.2188) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.3956 (1.4378) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 843.775, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.013) eta 1:20:32 loss 1.7985 (1.4069) acc 84.3750 (95.4688) lr 0.260000 +epoch: [32/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:31 loss 1.3286 (1.4361) acc 100.0000 (95.0000) lr 0.260000 +FPS@all 843.975, TIME@all 0.303 +epoch: [32/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.4344 (1.3838) acc 93.7500 (96.5625) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.5029 (1.4188) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 844.054, TIME@all 0.303 +epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.012) eta 1:20:33 loss 1.5059 (1.3754) acc 93.7500 (96.4062) lr 0.260000 +epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:32 loss 1.3600 (1.4100) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 843.893, TIME@all 0.303 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.2916 (1.3813) acc 96.8750 (96.0938) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:23 loss 1.5375 (1.4250) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 842.746, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.2976 (1.4035) acc 100.0000 (96.0938) lr 0.260000 +epoch: [33/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.2956 (1.4352) acc 100.0000 (94.9219) lr 0.260000 +FPS@all 842.779, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.3914 (1.3628) acc 93.7500 (96.7188) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.4482 (1.4207) acc 96.8750 (95.3906) lr 0.260000 +FPS@all 842.720, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:20:32 loss 1.5378 (1.3885) acc 90.6250 (95.6250) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:20:23 loss 1.3678 (1.4402) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 842.734, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:31 loss 1.3564 (1.3609) acc 96.8750 (96.7188) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4284 (1.4331) acc 96.8750 (94.4531) lr 0.260000 +FPS@all 842.902, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:32 loss 1.4260 (1.4160) acc 93.7500 (96.0938) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.5616 (1.4525) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 842.747, TIME@all 0.304 +epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:31 loss 1.2951 (1.3682) acc 96.8750 (97.1875) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4112 (1.4399) acc 93.7500 (94.6875) lr 0.260000 +FPS@all 842.894, TIME@all 0.304 +epoch: [33/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:20:30 loss 1.3132 (1.4014) acc 100.0000 (95.4688) lr 0.260000 +epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4165 (1.4387) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 843.127, TIME@all 0.304 +epoch: [34/350][20/50] time 0.305 (0.305) data 0.000 (0.011) eta 1:20:26 loss 1.4535 (1.4110) acc 96.8750 (94.3750) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.6118 (1.4344) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 842.664, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:20:26 loss 1.4155 (1.3867) acc 93.7500 (95.1562) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.5073 (1.4276) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 842.570, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:20:26 loss 1.6079 (1.4238) acc 90.6250 (94.5312) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.3945 (1.4345) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 842.610, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:26 loss 1.2972 (1.3779) acc 96.8750 (96.0938) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.4440 (1.4058) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 842.596, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:26 loss 1.2703 (1.4125) acc 100.0000 (95.4688) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.3298 (1.4398) acc 100.0000 (94.6875) lr 0.260000 +FPS@all 842.610, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:20:25 loss 1.4618 (1.4121) acc 87.5000 (94.8438) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:11 loss 1.7260 (1.4431) acc 84.3750 (94.3750) lr 0.260000 +FPS@all 842.787, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:20:25 loss 1.2716 (1.4210) acc 100.0000 (95.6250) lr 0.260000 +epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:11 loss 1.8364 (1.4641) acc 81.2500 (93.6719) lr 0.260000 +FPS@all 842.730, TIME@all 0.304 +epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:25 loss 1.3431 (1.3800) acc 93.7500 (96.0938) lr 0.260000 +epoch: [34/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.5976 (1.4302) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 842.996, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.4167 (1.3761) acc 96.8750 (96.0938) lr 0.260000 +epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:51 loss 1.5788 (1.4016) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 843.027, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.5644 (1.3769) acc 90.6250 (95.3125) lr 0.260000 +epoch: [35/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:50 loss 1.4647 (1.4222) acc 100.0000 (94.4531) lr 0.260000 +FPS@all 843.069, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.4607 (1.3885) acc 93.7500 (95.6250) lr 0.260000 +epoch: [35/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 1:19:50 loss 1.6069 (1.4099) acc 93.7500 (94.9219) lr 0.260000 +FPS@all 843.058, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.4376 (1.3777) acc 93.7500 (96.5625) lr 0.260000 +epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:49 loss 1.3996 (1.4042) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 843.254, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:58 loss 1.5132 (1.3841) acc 96.8750 (96.2500) lr 0.260000 +epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:51 loss 1.6493 (1.4118) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 843.050, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.3336 (1.3741) acc 96.8750 (95.6250) lr 0.260000 +epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:50 loss 1.3826 (1.3967) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 843.219, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.4977 (1.3960) acc 84.3750 (95.3125) lr 0.260000 +epoch: [35/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:50 loss 1.5070 (1.4029) acc 93.7500 (95.0000) lr 0.260000 +FPS@all 843.474, TIME@all 0.304 +epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:58 loss 1.3807 (1.3839) acc 100.0000 (95.3125) lr 0.260000 +epoch: [35/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:19:50 loss 1.4870 (1.4141) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 843.048, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:19:57 loss 1.6046 (1.4545) acc 87.5000 (93.1250) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:41 loss 1.3781 (1.4722) acc 96.8750 (92.5781) lr 0.260000 +FPS@all 842.341, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:19:56 loss 1.4084 (1.4327) acc 100.0000 (95.6250) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.2522 (1.4642) acc 100.0000 (94.4531) lr 0.260000 +FPS@all 842.398, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:19:56 loss 1.8281 (1.4579) acc 84.3750 (94.3750) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:41 loss 1.3764 (1.4585) acc 93.7500 (94.0625) lr 0.260000 +FPS@all 842.410, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.5766 (1.4145) acc 90.6250 (95.0000) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.3913 (1.4456) acc 96.8750 (94.1406) lr 0.260000 +FPS@all 842.355, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.6903 (1.3756) acc 81.2500 (96.4062) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.3317 (1.4380) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 842.343, TIME@all 0.304 +epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:19:55 loss 1.4633 (1.4032) acc 87.5000 (95.3125) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:40 loss 1.6214 (1.4449) acc 90.6250 (93.7500) lr 0.260000 +FPS@all 842.536, TIME@all 0.304 +epoch: [36/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:19:54 loss 1.4693 (1.4047) acc 93.7500 (95.6250) lr 0.260000 +epoch: [36/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.7239 (1.4646) acc 87.5000 (93.7500) lr 0.260000 +FPS@all 842.805, TIME@all 0.304 +epoch: [36/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.4292 (1.3865) acc 90.6250 (95.7812) lr 0.260000 +epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:40 loss 1.3963 (1.4119) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 842.485, TIME@all 0.304 +epoch: [37/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.3809 (1.3763) acc 96.8750 (96.0938) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:14 loss 1.5800 (1.4321) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 844.453, TIME@all 0.303 +epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:19:25 loss 1.5023 (1.4105) acc 87.5000 (94.0625) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:15 loss 1.3651 (1.4558) acc 96.8750 (93.2812) lr 0.260000 +FPS@all 844.347, TIME@all 0.303 +epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.5963 (1.4039) acc 93.7500 (95.4688) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.3469 (1.4579) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 844.389, TIME@all 0.303 +epoch: [37/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.5471 (1.3827) acc 90.6250 (95.9375) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.6800 (1.4578) acc 87.5000 (93.9844) lr 0.260000 +FPS@all 844.388, TIME@all 0.303 +epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.3834 (1.3827) acc 93.7500 (95.7812) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:19:15 loss 1.4566 (1.4524) acc 90.6250 (93.4375) lr 0.260000 +FPS@all 844.370, TIME@all 0.303 +epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:24 loss 1.3389 (1.3897) acc 100.0000 (96.4062) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:14 loss 1.3661 (1.4559) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 844.564, TIME@all 0.303 +epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:24 loss 1.4975 (1.3775) acc 93.7500 (95.9375) lr 0.260000 +epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:14 loss 1.2930 (1.4391) acc 100.0000 (94.3750) lr 0.260000 +FPS@all 844.536, TIME@all 0.303 +epoch: [37/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 1:19:21 loss 1.5539 (1.3801) acc 87.5000 (96.4062) lr 0.260000 +epoch: [37/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.4530 (1.4456) acc 93.7500 (94.8438) lr 0.260000 +FPS@all 844.729, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4846 (1.3458) acc 90.6250 (96.4062) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.3086 (1.3842) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 844.408, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 1:19:01 loss 1.3799 (1.3820) acc 96.8750 (95.3125) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:19:02 loss 1.4939 (1.4265) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 844.299, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4040 (1.3686) acc 96.8750 (96.7188) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.3801 (1.4058) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 844.340, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.3569 (1.3729) acc 96.8750 (96.4062) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.5525 (1.4126) acc 90.6250 (95.0000) lr 0.260000 +FPS@all 844.320, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 1:19:00 loss 1.4394 (1.3431) acc 96.8750 (97.3438) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:01 loss 1.2587 (1.3935) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 844.530, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4167 (1.3556) acc 96.8750 (96.2500) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.4004 (1.3987) acc 90.6250 (95.2344) lr 0.260000 +FPS@all 844.343, TIME@all 0.303 +epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.5171 (1.3479) acc 90.6250 (96.7188) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:01 loss 1.4844 (1.3842) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 844.481, TIME@all 0.303 +epoch: [38/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4400 (1.3629) acc 90.6250 (95.9375) lr 0.260000 +epoch: [38/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 1:19:01 loss 1.3534 (1.3905) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 844.672, TIME@all 0.303 +epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:19:12 loss 1.3539 (1.3648) acc 96.8750 (96.4062) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:18:57 loss 1.3486 (1.4230) acc 93.7500 (94.6875) lr 0.260000 +FPS@all 842.163, TIME@all 0.304 +epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:19:12 loss 1.4385 (1.3913) acc 96.8750 (95.7812) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:18:56 loss 1.4023 (1.4524) acc 90.6250 (93.9062) lr 0.260000 +FPS@all 842.225, TIME@all 0.304 +epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.3853 (1.3781) acc 96.8750 (95.9375) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:56 loss 1.3193 (1.4065) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 842.234, TIME@all 0.304 +epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.4674 (1.3877) acc 93.7500 (95.6250) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:57 loss 1.3463 (1.4210) acc 93.7500 (94.9219) lr 0.260000 +FPS@all 842.173, TIME@all 0.304 +epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.4147 (1.4315) acc 100.0000 (94.2188) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:57 loss 1.3317 (1.4443) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 842.172, TIME@all 0.304 +epoch: [39/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 1:19:11 loss 1.3255 (1.3796) acc 96.8750 (95.9375) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:55 loss 1.4420 (1.4289) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 842.378, TIME@all 0.304 +epoch: [39/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 1:19:11 loss 1.5030 (1.3760) acc 90.6250 (96.2500) lr 0.260000 +epoch: [39/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:18:56 loss 1.4192 (1.4320) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 842.515, TIME@all 0.304 +epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:19:11 loss 1.5416 (1.4052) acc 96.8750 (96.0938) lr 0.260000 +epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:56 loss 1.5239 (1.4596) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 842.323, TIME@all 0.304 +epoch: [40/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:19:00 loss 1.2576 (1.4050) acc 100.0000 (96.2500) lr 0.260000 +epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.4982 (1.4261) acc 90.6250 (95.2344) lr 0.260000 +FPS@all 840.898, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:18:59 loss 1.3215 (1.3954) acc 100.0000 (95.6250) lr 0.260000 +epoch: [40/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.5421 (1.4280) acc 90.6250 (94.6875) lr 0.260000 +FPS@all 840.967, TIME@all 0.304 +epoch: [40/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:18:59 loss 1.3232 (1.3708) acc 100.0000 (96.5625) lr 0.260000 +epoch: [40/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.4798 (1.4092) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 840.975, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:19:00 loss 1.4074 (1.3683) acc 93.7500 (95.6250) lr 0.260000 +epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:18:48 loss 1.4727 (1.4155) acc 90.6250 (94.6094) lr 0.260000 +FPS@all 840.922, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.012) eta 1:19:00 loss 1.3298 (1.3658) acc 96.8750 (96.4062) lr 0.260000 +epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 1:18:48 loss 1.3420 (1.4125) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 840.918, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:18:59 loss 1.3986 (1.3764) acc 93.7500 (96.4062) lr 0.260000 +epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 1:18:47 loss 1.5653 (1.3993) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 841.058, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.013) eta 1:18:58 loss 1.3725 (1.3840) acc 93.7500 (95.6250) lr 0.260000 +epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 1:18:47 loss 1.5646 (1.4220) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 841.118, TIME@all 0.304 +epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.012) eta 1:19:00 loss 1.3692 (1.4003) acc 96.8750 (95.9375) lr 0.260000 +epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.6419 (1.4250) acc 90.6250 (94.7656) lr 0.260000 +FPS@all 841.230, TIME@all 0.304 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:18:26 loss 1.4732 (1.4418) acc 93.7500 (95.6250) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.4219 (1.4474) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 843.876, TIME@all 0.303 +epoch: [41/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:18:27 loss 1.5240 (1.4289) acc 93.7500 (94.2188) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.4799 (1.4378) acc 90.6250 (94.2969) lr 0.260000 +FPS@all 843.845, TIME@all 0.303 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:18:26 loss 1.4274 (1.4163) acc 100.0000 (94.2188) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.5713 (1.4443) acc 90.6250 (93.9844) lr 0.260000 +FPS@all 843.865, TIME@all 0.303 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:26 loss 1.5023 (1.4226) acc 90.6250 (94.5312) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:19 loss 1.4060 (1.4299) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 844.034, TIME@all 0.303 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:27 loss 1.6324 (1.4363) acc 90.6250 (95.1562) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:20 loss 1.7515 (1.4644) acc 87.5000 (93.4375) lr 0.260000 +FPS@all 843.850, TIME@all 0.303 +epoch: [41/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:18:25 loss 1.7913 (1.4328) acc 81.2500 (93.7500) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.3862 (1.4307) acc 96.8750 (94.1406) lr 0.260000 +FPS@all 844.204, TIME@all 0.303 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:27 loss 1.3687 (1.4201) acc 96.8750 (94.8438) lr 0.260000 +epoch: [41/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:18:20 loss 1.4505 (1.4408) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 843.834, TIME@all 0.303 +epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:26 loss 1.3919 (1.4265) acc 96.8750 (95.0000) lr 0.260000 +epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:19 loss 1.3551 (1.4397) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 844.001, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.3690 (1.3649) acc 96.8750 (95.7812) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:59 loss 1.3367 (1.3923) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 844.188, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.7243 (1.4070) acc 87.5000 (95.6250) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:59 loss 1.3382 (1.4009) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 844.224, TIME@all 0.303 +epoch: [42/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 1:17:52 loss 1.5320 (1.3692) acc 96.8750 (96.5625) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:59 loss 1.3697 (1.4028) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 844.194, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.4798 (1.4165) acc 93.7500 (95.3125) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.2726 (1.4100) acc 100.0000 (94.9219) lr 0.260000 +FPS@all 844.377, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:50 loss 1.3823 (1.3713) acc 100.0000 (95.9375) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:58 loss 1.2971 (1.3939) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 844.623, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.5232 (1.3893) acc 87.5000 (94.6875) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.3591 (1.3996) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 844.241, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.4646 (1.3977) acc 93.7500 (95.6250) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.2645 (1.3895) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 844.231, TIME@all 0.303 +epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.5353 (1.3767) acc 93.7500 (95.3125) lr 0.260000 +epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.4745 (1.4077) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 844.415, TIME@all 0.303 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.3505 (1.3229) acc 93.7500 (96.8750) lr 0.260000 +epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.4131 (1.3743) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 842.503, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.4861 (1.3385) acc 90.6250 (96.7188) lr 0.260000 +epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:54 loss 1.7041 (1.3833) acc 87.5000 (95.7812) lr 0.260000 +FPS@all 842.432, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.6326 (1.3280) acc 84.3750 (96.2500) lr 0.260000 +epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:53 loss 1.5188 (1.3630) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 842.484, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.3970 (1.3364) acc 96.8750 (96.5625) lr 0.260000 +epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.4296 (1.3693) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 842.507, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.3655 (1.2987) acc 96.8750 (97.5000) lr 0.260000 +epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:54 loss 1.4523 (1.3554) acc 90.6250 (96.0156) lr 0.260000 +FPS@all 842.439, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:18:03 loss 1.4184 (1.3275) acc 93.7500 (97.0312) lr 0.260000 +epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:17:52 loss 1.3124 (1.3582) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 842.666, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:03 loss 1.5042 (1.3457) acc 90.6250 (97.5000) lr 0.260000 +epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:53 loss 1.6956 (1.3947) acc 78.1250 (95.8594) lr 0.260000 +FPS@all 842.614, TIME@all 0.304 +epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.4499 (1.3134) acc 96.8750 (97.3438) lr 0.260000 +epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.2931 (1.3483) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 842.764, TIME@all 0.304 +epoch: [44/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:17:45 loss 1.5203 (1.3899) acc 90.6250 (95.9375) lr 0.260000 +epoch: [44/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.3939 (1.4193) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 842.873, TIME@all 0.304 +epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:17:47 loss 1.4746 (1.3596) acc 87.5000 (96.2500) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.4934 (1.3994) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 842.814, TIME@all 0.304 +epoch: [44/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:17:46 loss 1.4012 (1.3652) acc 93.7500 (96.2500) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.2976 (1.3924) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 842.871, TIME@all 0.304 +epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:17:46 loss 1.3537 (1.3492) acc 93.7500 (96.2500) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:17:40 loss 1.2906 (1.3904) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 842.852, TIME@all 0.304 +epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:17:46 loss 1.2827 (1.3777) acc 96.8750 (96.0938) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.4168 (1.4054) acc 96.8750 (94.9219) lr 0.260000 +FPS@all 842.816, TIME@all 0.304 +epoch: [44/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:17:45 loss 1.3694 (1.3779) acc 96.8750 (96.2500) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:17:39 loss 1.2714 (1.3988) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 843.001, TIME@all 0.304 +epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:17:46 loss 1.3043 (1.3765) acc 100.0000 (95.6250) lr 0.260000 +epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:17:39 loss 1.4881 (1.4039) acc 87.5000 (94.9219) lr 0.260000 +FPS@all 842.952, TIME@all 0.304 +epoch: [44/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:17:47 loss 1.4765 (1.3700) acc 93.7500 (95.7812) lr 0.260000 +epoch: [44/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:17:40 loss 1.4077 (1.4033) acc 93.7500 (95.2344) lr 0.260000 +FPS@all 843.157, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:17:42 loss 1.2758 (1.3500) acc 100.0000 (96.8750) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:17:31 loss 1.3525 (1.3929) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 841.247, TIME@all 0.304 +epoch: [45/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:17:42 loss 1.5576 (1.3549) acc 93.7500 (97.1875) lr 0.260000 +epoch: [45/350][40/50] time 0.306 (0.305) data 0.001 (0.006) eta 1:17:30 loss 1.4492 (1.4208) acc 93.7500 (94.6875) lr 0.260000 +FPS@all 841.299, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:17:41 loss 1.5842 (1.3539) acc 87.5000 (96.0938) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.3953 (1.3915) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 841.291, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:17:41 loss 1.3760 (1.3457) acc 96.8750 (97.0312) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:31 loss 1.4133 (1.3739) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 841.280, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:17:41 loss 1.3082 (1.3692) acc 100.0000 (95.7812) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.4138 (1.3949) acc 96.8750 (94.9219) lr 0.260000 +FPS@all 841.394, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:17:42 loss 1.3156 (1.3498) acc 96.8750 (96.8750) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:31 loss 1.3889 (1.3869) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 841.214, TIME@all 0.304 +epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 1:17:41 loss 1.5638 (1.3643) acc 90.6250 (95.4688) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.5100 (1.4101) acc 87.5000 (94.2969) lr 0.260000 +FPS@all 841.456, TIME@all 0.304 +epoch: [45/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:17:38 loss 1.3572 (1.3411) acc 96.8750 (96.7188) lr 0.260000 +epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.3643 (1.3695) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 841.684, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:17:23 loss 1.6539 (1.4238) acc 90.6250 (94.5312) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.006) eta 1:17:15 loss 1.3770 (1.4462) acc 96.8750 (93.9844) lr 0.260000 +FPS@all 841.317, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:17:23 loss 1.4017 (1.3992) acc 96.8750 (95.4688) lr 0.260000 +epoch: [46/350][40/50] time 0.309 (0.305) data 0.000 (0.006) eta 1:17:15 loss 1.3529 (1.4435) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 841.363, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.5369 (1.3698) acc 87.5000 (96.2500) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:15 loss 1.4765 (1.4204) acc 96.8750 (94.8438) lr 0.260000 +FPS@all 841.342, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.3305 (1.3616) acc 100.0000 (96.2500) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:15 loss 1.3361 (1.3974) acc 96.8750 (94.9219) lr 0.260000 +FPS@all 841.351, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:17:22 loss 1.7191 (1.3910) acc 90.6250 (95.4688) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.001 (0.007) eta 1:17:14 loss 1.3428 (1.4436) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 841.541, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.6737 (1.3984) acc 84.3750 (95.1562) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.001 (0.007) eta 1:17:15 loss 1.4671 (1.4560) acc 90.6250 (93.2031) lr 0.260000 +FPS@all 841.319, TIME@all 0.304 +epoch: [46/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:17:22 loss 1.5724 (1.3709) acc 90.6250 (96.4062) lr 0.260000 +epoch: [46/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:17:14 loss 1.4607 (1.4243) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 841.702, TIME@all 0.304 +epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.4585 (1.3717) acc 93.7500 (96.8750) lr 0.260000 +epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:14 loss 1.5116 (1.4253) acc 90.6250 (94.6875) lr 0.260000 +FPS@all 841.459, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:16:55 loss 1.4941 (1.3669) acc 90.6250 (96.7188) lr 0.260000 +epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:16:52 loss 1.3977 (1.4263) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 842.742, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:56 loss 1.6213 (1.3811) acc 84.3750 (95.6250) lr 0.260000 +epoch: [47/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.2942 (1.4394) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 842.745, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:16:55 loss 1.5742 (1.3783) acc 87.5000 (96.4062) lr 0.260000 +epoch: [47/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:16:52 loss 1.4398 (1.4173) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 842.669, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:16:54 loss 1.7294 (1.3950) acc 81.2500 (95.9375) lr 0.260000 +epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:51 loss 1.5283 (1.4531) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 842.887, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:16:55 loss 1.6822 (1.3763) acc 87.5000 (95.6250) lr 0.260000 +epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:51 loss 1.4442 (1.4378) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 842.833, TIME@all 0.304 +epoch: [47/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:54 loss 1.5721 (1.3950) acc 93.7500 (95.4688) lr 0.260000 +epoch: [47/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 1:16:51 loss 1.3653 (1.4449) acc 96.8750 (93.7500) lr 0.260000 +FPS@all 843.030, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:55 loss 1.6564 (1.4090) acc 90.6250 (95.0000) lr 0.260000 +epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.3278 (1.4365) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 842.694, TIME@all 0.304 +epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:55 loss 1.4687 (1.4010) acc 100.0000 (95.0000) lr 0.260000 +epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.3657 (1.4330) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 842.690, TIME@all 0.304 +epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:16:43 loss 1.6637 (1.3769) acc 90.6250 (96.0938) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:16:34 loss 1.4681 (1.4196) acc 96.8750 (94.6094) lr 0.260000 +FPS@all 842.709, TIME@all 0.304 +epoch: [48/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:16:43 loss 1.3808 (1.3219) acc 100.0000 (96.5625) lr 0.260000 +epoch: [48/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.4481 (1.3912) acc 87.5000 (94.2969) lr 0.260000 +FPS@all 842.756, TIME@all 0.304 +epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:44 loss 1.5695 (1.3755) acc 96.8750 (97.1875) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.3970 (1.4171) acc 93.7500 (94.8438) lr 0.260000 +FPS@all 842.706, TIME@all 0.304 +epoch: [48/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:16:44 loss 1.5247 (1.3576) acc 90.6250 (96.5625) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.5772 (1.4216) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 842.750, TIME@all 0.304 +epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:42 loss 1.5281 (1.3592) acc 90.6250 (95.9375) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.4027 (1.4067) acc 100.0000 (94.9219) lr 0.260000 +FPS@all 842.941, TIME@all 0.304 +epoch: [48/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:16:43 loss 1.6272 (1.3828) acc 90.6250 (95.9375) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 1:16:33 loss 1.4028 (1.4183) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 843.062, TIME@all 0.304 +epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:43 loss 1.8314 (1.3917) acc 78.1250 (95.0000) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.5794 (1.4468) acc 81.2500 (93.5156) lr 0.260000 +FPS@all 842.866, TIME@all 0.304 +epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:44 loss 1.4955 (1.3893) acc 93.7500 (96.4062) lr 0.260000 +epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.2927 (1.4312) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 842.708, TIME@all 0.304 +epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:16:20 loss 1.4345 (1.3489) acc 96.8750 (96.2500) lr 0.260000 +epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:16:20 loss 1.5156 (1.3939) acc 90.6250 (95.1562) lr 0.260000 +FPS@all 843.088, TIME@all 0.304 +epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.3402 (1.3384) acc 100.0000 (98.1250) lr 0.260000 +epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.4750 (1.3923) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 843.120, TIME@all 0.304 +epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.3404 (1.3462) acc 96.8750 (96.8750) lr 0.260000 +epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.2914 (1.3886) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 843.101, TIME@all 0.304 +epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.4825 (1.3539) acc 93.7500 (96.4062) lr 0.260000 +epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.5743 (1.3934) acc 90.6250 (95.3125) lr 0.260000 +FPS@all 843.302, TIME@all 0.304 +epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.3827 (1.3453) acc 93.7500 (96.5625) lr 0.260000 +epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.3679 (1.3691) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 843.108, TIME@all 0.304 +epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.5449 (1.3847) acc 90.6250 (95.4688) lr 0.260000 +epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.4562 (1.4327) acc 96.8750 (93.7500) lr 0.260000 +FPS@all 843.095, TIME@all 0.304 +epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.4416 (1.3709) acc 96.8750 (96.0938) lr 0.260000 +epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.6143 (1.4155) acc 90.6250 (94.2969) lr 0.260000 +FPS@all 843.216, TIME@all 0.304 +epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:21 loss 1.4899 (1.3760) acc 90.6250 (95.1562) lr 0.260000 +epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.4687 (1.3896) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 843.407, TIME@all 0.304 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4668 (1.3481) acc 90.6250 (96.4062) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.2377 (1.3754) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 843.535, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4682 (1.3419) acc 93.7500 (97.8125) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.3581 (1.3561) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 843.624, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.3858 (1.3712) acc 90.6250 (96.4062) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.2464 (1.3669) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 843.573, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.3636 (1.3229) acc 100.0000 (97.6562) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.4016 (1.3488) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 843.543, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4267 (1.3426) acc 93.7500 (96.7188) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.5222 (1.3618) acc 90.6250 (96.0156) lr 0.260000 +FPS@all 843.533, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:16:13 loss 1.3286 (1.3620) acc 100.0000 (96.5625) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 1:15:58 loss 1.2813 (1.3754) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 843.758, TIME@all 0.303 +epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:13 loss 1.3820 (1.3516) acc 100.0000 (96.0938) lr 0.260000 +epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.2784 (1.3839) acc 100.0000 (95.4688) lr 0.260000 +FPS@all 843.680, TIME@all 0.303 +epoch: [50/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:16:15 loss 1.6284 (1.3816) acc 87.5000 (95.4688) lr 0.260000 +epoch: [50/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.3628 (1.4008) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 843.836, TIME@all 0.303 +epoch: [51/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:16:08 loss 1.4294 (1.3719) acc 93.7500 (95.7812) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.3863 (1.3848) acc 90.6250 (95.3125) lr 0.260000 +FPS@all 842.879, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.4116 (1.3406) acc 96.8750 (95.9375) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.4322 (1.3785) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 842.831, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.3801 (1.3339) acc 100.0000 (97.5000) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.001 (0.006) eta 1:15:51 loss 1.4403 (1.3734) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 842.893, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.4574 (1.3464) acc 96.8750 (97.3438) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.4311 (1.3841) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 842.874, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.5432 (1.3577) acc 87.5000 (96.5625) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.5316 (1.3790) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 842.876, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.3982 (1.3600) acc 100.0000 (96.0938) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:15:50 loss 1.4276 (1.3885) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 843.062, TIME@all 0.304 +epoch: [51/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:16:08 loss 1.5042 (1.3501) acc 93.7500 (95.7812) lr 0.260000 +epoch: [51/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 1:15:50 loss 1.3474 (1.3859) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 843.283, TIME@all 0.304 +epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.2923 (1.3303) acc 100.0000 (97.3438) lr 0.260000 +epoch: [51/350][40/50] time 0.297 (0.304) data 0.001 (0.007) eta 1:15:50 loss 1.5538 (1.3658) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 842.996, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.5025 (1.3559) acc 90.6250 (96.2500) lr 0.260000 +epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:31 loss 1.4122 (1.3741) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 843.277, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:42 loss 1.3861 (1.3450) acc 93.7500 (95.9375) lr 0.260000 +epoch: [52/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4924 (1.3935) acc 93.7500 (95.0000) lr 0.260000 +FPS@all 843.326, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.3561 (1.3438) acc 96.8750 (95.7812) lr 0.260000 +epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:30 loss 1.3935 (1.3660) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 843.389, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4355 (1.3560) acc 93.7500 (96.2500) lr 0.260000 +epoch: [52/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4628 (1.3879) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 843.328, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.4102 (1.3808) acc 93.7500 (95.7812) lr 0.260000 +epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:30 loss 1.4888 (1.3926) acc 90.6250 (95.1562) lr 0.260000 +FPS@all 843.300, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4135 (1.3505) acc 93.7500 (95.6250) lr 0.260000 +epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:15:29 loss 1.5182 (1.3806) acc 87.5000 (95.2344) lr 0.260000 +FPS@all 843.485, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4932 (1.3464) acc 90.6250 (96.8750) lr 0.260000 +epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4908 (1.3912) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 843.443, TIME@all 0.304 +epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:41 loss 1.5703 (1.3512) acc 87.5000 (96.2500) lr 0.260000 +epoch: [52/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:15:29 loss 1.4527 (1.3794) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 843.610, TIME@all 0.303 +epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:15:31 loss 1.2600 (1.3451) acc 100.0000 (96.2500) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.4386 (1.3544) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 843.322, TIME@all 0.304 +epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.2617 (1.3504) acc 100.0000 (96.7188) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:16 loss 1.3262 (1.3588) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 843.428, TIME@all 0.304 +epoch: [53/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3459 (1.3519) acc 96.8750 (96.7188) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:16 loss 1.3717 (1.3731) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 843.365, TIME@all 0.304 +epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:32 loss 1.3196 (1.3398) acc 96.8750 (96.8750) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.5282 (1.3740) acc 90.6250 (96.0156) lr 0.260000 +FPS@all 843.346, TIME@all 0.304 +epoch: [53/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:30 loss 1.4106 (1.3634) acc 93.7500 (95.3125) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.007) eta 1:15:16 loss 1.3932 (1.3710) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 843.546, TIME@all 0.303 +epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3302 (1.3481) acc 96.8750 (96.2500) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.3857 (1.3613) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 843.331, TIME@all 0.304 +epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:15:31 loss 1.3498 (1.3419) acc 93.7500 (96.8750) lr 0.260000 +epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.007) eta 1:15:16 loss 1.5431 (1.3683) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 843.480, TIME@all 0.304 +epoch: [53/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3679 (1.3353) acc 93.7500 (96.8750) lr 0.260000 +epoch: [53/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.3967 (1.3707) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 843.681, TIME@all 0.303 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:15:10 loss 1.3664 (1.3004) acc 96.8750 (97.9688) lr 0.260000 +epoch: [54/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.2155 (1.3411) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 842.744, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:15:10 loss 1.2573 (1.3227) acc 100.0000 (96.5625) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.2661 (1.3564) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 842.745, TIME@all 0.304 +epoch: [54/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 1:15:11 loss 1.3846 (1.3395) acc 96.8750 (97.0312) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:02 loss 1.2787 (1.3624) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 842.665, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:10 loss 1.3902 (1.3296) acc 93.7500 (97.1875) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:02 loss 1.3812 (1.3678) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 842.692, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:11 loss 1.3139 (1.3080) acc 96.8750 (97.3438) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.3454 (1.3469) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 842.678, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:10 loss 1.2916 (1.3188) acc 96.8750 (96.5625) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:01 loss 1.4392 (1.3537) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 842.844, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:13 loss 1.4681 (1.3321) acc 90.6250 (97.5000) lr 0.260000 +epoch: [54/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:15:04 loss 1.3051 (1.3581) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 842.874, TIME@all 0.304 +epoch: [54/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 1:15:10 loss 1.2272 (1.3215) acc 100.0000 (97.9688) lr 0.260000 +epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:01 loss 1.3147 (1.3649) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 842.880, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:09 loss 1.3413 (1.3454) acc 93.7500 (95.7812) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.6728 (1.3724) acc 90.6250 (95.2344) lr 0.260000 +FPS@all 841.664, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:09 loss 1.4178 (1.3512) acc 96.8750 (96.2500) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.4615 (1.3747) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 841.718, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:07 loss 1.5324 (1.3393) acc 93.7500 (96.8750) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:57 loss 1.4710 (1.3502) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 841.783, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 1:15:09 loss 1.4522 (1.3468) acc 90.6250 (96.7188) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.4574 (1.3604) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 841.695, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:09 loss 1.7154 (1.3533) acc 90.6250 (97.1875) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:58 loss 1.4411 (1.3743) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 841.646, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:08 loss 1.4800 (1.3501) acc 93.7500 (96.7188) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.4767 (1.3454) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 841.810, TIME@all 0.304 +epoch: [55/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:15:08 loss 1.5136 (1.3481) acc 90.6250 (96.5625) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.5412 (1.3524) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 841.861, TIME@all 0.304 +epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:07 loss 1.5582 (1.3241) acc 96.8750 (97.1875) lr 0.260000 +epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.3947 (1.3474) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 842.116, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2974 (1.3412) acc 100.0000 (96.2500) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.2866 (1.3650) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 842.501, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:48 loss 1.3311 (1.3352) acc 93.7500 (96.7188) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4252 (1.3733) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 842.520, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2290 (1.3495) acc 100.0000 (95.7812) lr 0.260000 +epoch: [56/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4373 (1.3817) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 842.407, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2738 (1.3499) acc 100.0000 (97.1875) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.3066 (1.3605) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 842.468, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:14:48 loss 1.2391 (1.3109) acc 100.0000 (97.6562) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:38 loss 1.3469 (1.3458) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 842.645, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2903 (1.3422) acc 96.8750 (96.8750) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4900 (1.3550) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 842.471, TIME@all 0.304 +epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:14:49 loss 1.2782 (1.3303) acc 100.0000 (96.8750) lr 0.260000 +epoch: [56/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:38 loss 1.3370 (1.3529) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 842.579, TIME@all 0.304 +epoch: [56/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.3400 (1.3395) acc 96.8750 (96.4062) lr 0.260000 +epoch: [56/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.5472 (1.3847) acc 90.6250 (95.2344) lr 0.260000 +FPS@all 842.729, TIME@all 0.304 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.011) eta 1:14:18 loss 1.2647 (1.3287) acc 100.0000 (97.3438) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.3021 (1.3641) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 845.171, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 1:14:17 loss 1.4495 (1.3449) acc 93.7500 (95.7812) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:04 loss 1.4306 (1.3709) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 845.250, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:14:18 loss 1.4836 (1.3410) acc 90.6250 (95.9375) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 1:14:05 loss 1.2225 (1.3727) acc 100.0000 (95.3906) lr 0.260000 +FPS@all 845.195, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.3485 (1.3326) acc 96.8750 (97.0312) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.4996 (1.3601) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 845.188, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.4348 (1.3139) acc 90.6250 (96.8750) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.5361 (1.3592) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 845.190, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:14:16 loss 1.2825 (1.3057) acc 100.0000 (97.1875) lr 0.260000 +epoch: [57/350][40/50] time 0.300 (0.303) data 0.001 (0.007) eta 1:14:04 loss 1.3572 (1.3613) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 845.385, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 1:14:16 loss 1.3932 (1.3411) acc 100.0000 (97.0312) lr 0.260000 +epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 1:14:04 loss 1.3425 (1.3682) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 845.356, TIME@all 0.303 +epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.4809 (1.3093) acc 87.5000 (97.0312) lr 0.260000 +epoch: [57/350][40/50] time 0.303 (0.303) data 0.001 (0.006) eta 1:14:04 loss 1.4408 (1.3593) acc 93.7500 (95.8594) lr 0.260000 +FPS@all 845.540, TIME@all 0.303 +epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.4080 (1.3206) acc 100.0000 (96.7188) lr 0.260000 +epoch: [58/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.2940 (1.3406) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 842.820, TIME@all 0.304 +epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.2434 (1.3120) acc 100.0000 (97.1875) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.3424 (1.3531) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 842.803, TIME@all 0.304 +epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 1:14:22 loss 1.4265 (1.3244) acc 96.8750 (97.0312) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:14:07 loss 1.3450 (1.3476) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 842.742, TIME@all 0.304 +epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.014) eta 1:14:21 loss 1.3292 (1.3288) acc 96.8750 (95.9375) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.4119 (1.3536) acc 93.7500 (95.7812) lr 0.260000 +FPS@all 842.915, TIME@all 0.304 +epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.014) eta 1:14:20 loss 1.3882 (1.3274) acc 90.6250 (96.5625) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:06 loss 1.3441 (1.3380) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 842.962, TIME@all 0.304 +epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.4081 (1.3096) acc 96.8750 (97.0312) lr 0.260000 +epoch: [58/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.3607 (1.3365) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 842.779, TIME@all 0.304 +epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:22 loss 1.3045 (1.3249) acc 100.0000 (96.8750) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.2861 (1.3527) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 842.754, TIME@all 0.304 +epoch: [58/350][20/50] time 0.309 (0.305) data 0.001 (0.013) eta 1:14:20 loss 1.3346 (1.3157) acc 96.8750 (96.4062) lr 0.260000 +epoch: [58/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:14:06 loss 1.2801 (1.3490) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 843.163, TIME@all 0.304 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4228 (1.3389) acc 93.7500 (97.3438) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3923 (1.3584) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 843.461, TIME@all 0.304 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4597 (1.3366) acc 96.8750 (97.0312) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.2432 (1.3569) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 843.422, TIME@all 0.304 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4448 (1.3521) acc 96.8750 (96.8750) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3688 (1.3622) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 843.359, TIME@all 0.304 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4144 (1.3251) acc 96.8750 (96.7188) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3814 (1.3450) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 843.393, TIME@all 0.304 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:13:56 loss 1.4230 (1.3192) acc 100.0000 (97.0312) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:13:44 loss 1.4836 (1.3705) acc 90.6250 (95.8594) lr 0.260000 +FPS@all 843.551, TIME@all 0.303 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:13:56 loss 1.4001 (1.3330) acc 93.7500 (96.8750) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:13:44 loss 1.4199 (1.3518) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 843.539, TIME@all 0.303 +epoch: [59/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:56 loss 1.4002 (1.3248) acc 93.7500 (97.1875) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.2870 (1.3363) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 843.732, TIME@all 0.303 +epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:13:57 loss 1.4090 (1.3478) acc 100.0000 (97.1875) lr 0.260000 +epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:13:45 loss 1.4894 (1.3630) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 843.397, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.3193 (1.3508) acc 96.8750 (96.0938) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.2403 (1.3592) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 843.250, TIME@all 0.304 +epoch: [60/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:13:29 loss 1.4471 (1.3171) acc 93.7500 (97.1875) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:13:32 loss 1.2966 (1.3417) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 843.338, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:13:30 loss 1.2352 (1.3504) acc 100.0000 (96.2500) lr 0.260000 +epoch: [60/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:13:33 loss 1.2901 (1.3574) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 843.202, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.3026 (1.3251) acc 100.0000 (97.1875) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.3674 (1.3429) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 843.215, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.2748 (1.3512) acc 96.8750 (96.4062) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.4112 (1.3705) acc 90.6250 (95.5469) lr 0.260000 +FPS@all 843.378, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.4601 (1.3453) acc 90.6250 (96.5625) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.5230 (1.3682) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 843.217, TIME@all 0.304 +epoch: [60/350][20/50] time 0.305 (0.303) data 0.000 (0.014) eta 1:13:29 loss 1.2677 (1.3572) acc 96.8750 (96.0938) lr 0.260000 +epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.3091 (1.3582) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 843.422, TIME@all 0.304 +epoch: [60/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:13:29 loss 1.3276 (1.3614) acc 93.7500 (95.3125) lr 0.260000 +epoch: [60/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.2973 (1.3666) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 843.675, TIME@all 0.303 +epoch: [61/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2977 (1.2904) acc 93.7500 (97.0312) lr 0.260000 +epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.4225 (1.3346) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 843.274, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:25 loss 1.3211 (1.3150) acc 93.7500 (97.5000) lr 0.260000 +epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.2626 (1.3450) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 843.215, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2201 (1.3379) acc 100.0000 (96.8750) lr 0.260000 +epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:17 loss 1.3599 (1.3795) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 843.248, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2702 (1.2961) acc 96.8750 (96.8750) lr 0.260000 +epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:17 loss 1.2461 (1.3210) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 843.233, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:24 loss 1.3587 (1.3580) acc 100.0000 (96.4062) lr 0.260000 +epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:16 loss 1.3215 (1.3481) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 843.430, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.1878 (1.3174) acc 100.0000 (97.5000) lr 0.260000 +epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:16 loss 1.2605 (1.3268) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 843.375, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2797 (1.3195) acc 100.0000 (96.7188) lr 0.260000 +epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.3912 (1.3417) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 843.234, TIME@all 0.304 +epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:25 loss 1.2732 (1.2843) acc 100.0000 (98.5938) lr 0.260000 +epoch: [61/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.3154 (1.3203) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 843.593, TIME@all 0.303 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:13:12 loss 1.6327 (1.3188) acc 90.6250 (96.7188) lr 0.260000 +epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:13:06 loss 1.3067 (1.3650) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 842.696, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:13:13 loss 1.5963 (1.3694) acc 87.5000 (95.0000) lr 0.260000 +epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:07 loss 1.3803 (1.3873) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 842.613, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.7135 (1.3618) acc 87.5000 (96.0938) lr 0.260000 +epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:13:06 loss 1.4224 (1.4043) acc 96.8750 (95.0781) lr 0.260000 +FPS@all 842.719, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:12 loss 1.4728 (1.3520) acc 93.7500 (95.7812) lr 0.260000 +epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.4655 (1.3760) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 842.800, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.6463 (1.3574) acc 93.7500 (95.7812) lr 0.260000 +epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.4464 (1.3894) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 842.668, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.8071 (1.3706) acc 81.2500 (95.6250) lr 0.260000 +epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.5875 (1.3931) acc 87.5000 (95.0000) lr 0.260000 +FPS@all 842.668, TIME@all 0.304 +epoch: [62/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:13:12 loss 1.5690 (1.3385) acc 84.3750 (96.8750) lr 0.260000 +epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:05 loss 1.4180 (1.3784) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 842.823, TIME@all 0.304 +epoch: [62/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:10 loss 1.5478 (1.3385) acc 90.6250 (96.8750) lr 0.260000 +epoch: [62/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.3710 (1.3834) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 842.992, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 1:13:02 loss 1.6306 (1.4578) acc 90.6250 (93.5938) lr 0.260000 +epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.3748 (1.4955) acc 96.8750 (92.9688) lr 0.260000 +FPS@all 842.685, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.5835 (1.4247) acc 90.6250 (94.8438) lr 0.260000 +epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.7099 (1.4941) acc 84.3750 (92.7344) lr 0.260000 +FPS@all 842.736, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.7069 (1.4218) acc 87.5000 (95.1562) lr 0.260000 +epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.3751 (1.5016) acc 96.8750 (92.4219) lr 0.260000 +FPS@all 842.721, TIME@all 0.304 +epoch: [63/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 1:13:01 loss 1.6981 (1.4126) acc 90.6250 (95.7812) lr 0.260000 +epoch: [63/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:12:49 loss 1.4541 (1.4850) acc 93.7500 (93.2031) lr 0.260000 +FPS@all 842.747, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.6007 (1.4060) acc 90.6250 (96.2500) lr 0.260000 +epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.4705 (1.4874) acc 93.7500 (93.8281) lr 0.260000 +FPS@all 842.696, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:01 loss 1.4483 (1.4269) acc 96.8750 (95.4688) lr 0.260000 +epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:49 loss 1.5096 (1.4915) acc 93.7500 (94.0625) lr 0.260000 +FPS@all 842.820, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:01 loss 1.6165 (1.4250) acc 93.7500 (95.3125) lr 0.260000 +epoch: [63/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.4450 (1.4697) acc 93.7500 (94.0625) lr 0.260000 +FPS@all 843.040, TIME@all 0.304 +epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:13:01 loss 1.5048 (1.4109) acc 96.8750 (94.6875) lr 0.260000 +epoch: [63/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:12:49 loss 1.5889 (1.4873) acc 93.7500 (92.7344) lr 0.260000 +FPS@all 842.879, TIME@all 0.304 +epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:12:31 loss 1.8600 (1.4437) acc 81.2500 (94.2188) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:12:32 loss 1.3344 (1.4269) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 842.925, TIME@all 0.304 +epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.7978 (1.4431) acc 81.2500 (93.5938) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3815 (1.4709) acc 96.8750 (93.0469) lr 0.260000 +FPS@all 842.982, TIME@all 0.304 +epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.6890 (1.4357) acc 87.5000 (94.3750) lr 0.260000 +epoch: [64/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3250 (1.4359) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 842.971, TIME@all 0.304 +epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 2.0128 (1.4416) acc 71.8750 (94.5312) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3284 (1.4604) acc 96.8750 (93.5938) lr 0.260000 +FPS@all 842.925, TIME@all 0.304 +epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.7298 (1.4001) acc 90.6250 (95.6250) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:31 loss 1.4155 (1.4284) acc 100.0000 (95.0000) lr 0.260000 +FPS@all 843.070, TIME@all 0.304 +epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.6356 (1.4208) acc 90.6250 (94.5312) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:31 loss 1.3680 (1.4530) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 843.118, TIME@all 0.304 +epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:29 loss 1.6852 (1.4209) acc 90.6250 (95.1562) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:12:31 loss 1.5108 (1.4438) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 843.267, TIME@all 0.304 +epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:12:30 loss 2.0156 (1.4235) acc 78.1250 (95.3125) lr 0.260000 +epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.4068 (1.4408) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 842.934, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:17 loss 1.4271 (1.3436) acc 93.7500 (97.8125) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.3954 (1.3674) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 843.054, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:18 loss 1.4852 (1.3533) acc 96.8750 (96.4062) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4411 (1.3829) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 843.012, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6940 (1.3428) acc 87.5000 (96.8750) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.3722 (1.3733) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 842.895, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6812 (1.3573) acc 87.5000 (95.9375) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.3823 (1.3609) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 842.933, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6520 (1.3726) acc 93.7500 (96.0938) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.4473 (1.3941) acc 87.5000 (94.9219) lr 0.260000 +FPS@all 842.931, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:12:18 loss 1.4425 (1.3580) acc 93.7500 (95.7812) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:12:18 loss 1.4277 (1.3738) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 843.119, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6428 (1.3625) acc 90.6250 (96.8750) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4026 (1.3752) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 843.043, TIME@all 0.304 +epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:18 loss 1.4454 (1.3404) acc 93.7500 (97.0312) lr 0.260000 +epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4990 (1.3608) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 843.277, TIME@all 0.304 +epoch: [66/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:12:09 loss 1.2669 (1.3142) acc 100.0000 (97.0312) lr 0.260000 +epoch: [66/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.3008 (1.3549) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 841.919, TIME@all 0.304 +epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:12:08 loss 1.4964 (1.2883) acc 87.5000 (97.8125) lr 0.260000 +epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.2119 (1.3021) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 841.978, TIME@all 0.304 +epoch: [66/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:12:09 loss 1.4019 (1.2943) acc 93.7500 (98.2812) lr 0.260000 +epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.2801 (1.3310) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 841.872, TIME@all 0.304 +epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.6525 (1.3076) acc 90.6250 (96.8750) lr 0.260000 +epoch: [66/350][40/50] time 0.308 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.3389 (1.3433) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 841.934, TIME@all 0.304 +epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.3781 (1.3221) acc 93.7500 (96.5625) lr 0.260000 +epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:12:05 loss 1.2421 (1.3417) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 842.101, TIME@all 0.304 +epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.2878 (1.2871) acc 100.0000 (98.2812) lr 0.260000 +epoch: [66/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.2744 (1.3262) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 841.916, TIME@all 0.304 +epoch: [66/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:12:08 loss 1.2692 (1.2847) acc 100.0000 (98.1250) lr 0.260000 +epoch: [66/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.2470 (1.3130) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 842.238, TIME@all 0.304 +epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.4780 (1.3149) acc 96.8750 (97.8125) lr 0.260000 +epoch: [66/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 1:12:05 loss 1.3510 (1.3300) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.043, TIME@all 0.304 +epoch: [67/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 1:11:56 loss 1.6315 (1.3038) acc 90.6250 (97.8125) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.4039 (1.3411) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 842.706, TIME@all 0.304 +epoch: [67/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:12:00 loss 1.4870 (1.3153) acc 96.8750 (97.5000) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.3249 (1.3300) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 842.750, TIME@all 0.304 +epoch: [67/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 1:11:57 loss 1.5381 (1.3011) acc 87.5000 (97.0312) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.3782 (1.3322) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 842.747, TIME@all 0.304 +epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:57 loss 1.3772 (1.2872) acc 90.6250 (96.5625) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3688 (1.3271) acc 87.5000 (95.9375) lr 0.260000 +FPS@all 842.713, TIME@all 0.304 +epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:57 loss 1.4402 (1.2910) acc 93.7500 (97.8125) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3249 (1.3203) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 842.726, TIME@all 0.304 +epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.5720 (1.3029) acc 87.5000 (96.7188) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:46 loss 1.4102 (1.3204) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 842.913, TIME@all 0.304 +epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.4870 (1.3240) acc 90.6250 (95.6250) lr 0.260000 +epoch: [67/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3048 (1.3424) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 843.057, TIME@all 0.304 +epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.4612 (1.3030) acc 90.6250 (96.8750) lr 0.260000 +epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3369 (1.3237) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 842.842, TIME@all 0.304 +epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.2848 (1.2666) acc 96.8750 (97.9688) lr 0.260000 +epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.4895 (1.3024) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 842.769, TIME@all 0.304 +epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.3383 (1.2770) acc 96.8750 (98.4375) lr 0.260000 +epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.2673 (1.3096) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 842.800, TIME@all 0.304 +epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.2791 (1.2641) acc 96.8750 (98.2812) lr 0.260000 +epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.2372 (1.2824) acc 100.0000 (97.9688) lr 0.260000 +FPS@all 842.718, TIME@all 0.304 +epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:37 loss 1.2076 (1.2802) acc 100.0000 (98.4375) lr 0.260000 +epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.2608 (1.3206) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.767, TIME@all 0.304 +epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:36 loss 1.4035 (1.2842) acc 96.8750 (97.8125) lr 0.260000 +epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:32 loss 1.2223 (1.3043) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 842.940, TIME@all 0.304 +epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:36 loss 1.1997 (1.2662) acc 100.0000 (97.9688) lr 0.260000 +epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:32 loss 1.3210 (1.3103) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.899, TIME@all 0.304 +epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:37 loss 1.3303 (1.2908) acc 96.8750 (97.9688) lr 0.260000 +epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.4232 (1.3196) acc 90.6250 (97.4219) lr 0.260000 +FPS@all 842.755, TIME@all 0.304 +epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:39 loss 1.2603 (1.2626) acc 100.0000 (97.6562) lr 0.260000 +epoch: [68/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.5740 (1.3183) acc 87.5000 (96.5625) lr 0.260000 +FPS@all 842.892, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2599 (1.2779) acc 100.0000 (98.4375) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3795 (1.3087) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.671, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.4603 (1.2825) acc 93.7500 (97.3438) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3696 (1.3033) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.617, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2314 (1.2696) acc 100.0000 (98.9062) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.2859 (1.3046) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 842.718, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.3061 (1.2763) acc 96.8750 (97.8125) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3492 (1.3122) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.643, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2660 (1.2790) acc 100.0000 (98.1250) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.4566 (1.2915) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 842.670, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:14 loss 1.2204 (1.2810) acc 100.0000 (98.2812) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 1:11:21 loss 1.3764 (1.3303) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.827, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:14 loss 1.3620 (1.2916) acc 96.8750 (97.5000) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 1:11:21 loss 1.4330 (1.3189) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 842.856, TIME@all 0.304 +epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:14 loss 1.5021 (1.3063) acc 93.7500 (97.0312) lr 0.260000 +epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.2842 (1.3214) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 843.068, TIME@all 0.304 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.6224 (1.2752) acc 87.5000 (97.9688) lr 0.260000 +epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:29 loss 1.2994 (1.3150) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 839.053, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.3359 (1.2756) acc 93.7500 (97.8125) lr 0.260000 +epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:29 loss 1.2923 (1.3133) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 839.021, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.3040 (1.2631) acc 93.7500 (98.7500) lr 0.260000 +epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:28 loss 1.4032 (1.3087) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 839.090, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:39 loss 1.4123 (1.3330) acc 93.7500 (97.0312) lr 0.260000 +epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.2989 (1.3425) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 839.037, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.014) eta 1:11:38 loss 1.3629 (1.2684) acc 96.8750 (98.1250) lr 0.260000 +epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:27 loss 1.3690 (1.2976) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 839.239, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:39 loss 1.5699 (1.2876) acc 90.6250 (97.3438) lr 0.260000 +epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 1:11:29 loss 1.3177 (1.3113) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 839.049, TIME@all 0.305 +epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:38 loss 1.4117 (1.2602) acc 93.7500 (98.1250) lr 0.260000 +epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.3073 (1.2949) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 839.175, TIME@all 0.305 +epoch: [70/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 1:11:40 loss 1.2623 (1.2744) acc 96.8750 (98.4375) lr 0.260000 +epoch: [70/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.5246 (1.3314) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 839.388, TIME@all 0.305 +epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.011) eta 1:10:57 loss 1.4522 (1.4024) acc 90.6250 (94.6875) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.3316 (1.4615) acc 100.0000 (93.9062) lr 0.260000 +FPS@all 843.159, TIME@all 0.304 +epoch: [71/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.6315 (1.3950) acc 90.6250 (94.6875) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 1:10:46 loss 1.6204 (1.4642) acc 90.6250 (93.3594) lr 0.260000 +FPS@all 843.216, TIME@all 0.304 +epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5799 (1.4491) acc 90.6250 (94.0625) lr 0.260000 +epoch: [71/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.2984 (1.4894) acc 100.0000 (94.2188) lr 0.260000 +FPS@all 843.226, TIME@all 0.304 +epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:58 loss 1.4175 (1.3849) acc 96.8750 (95.6250) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.6282 (1.4559) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 843.190, TIME@all 0.304 +epoch: [71/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:10:56 loss 1.4903 (1.3910) acc 96.8750 (95.6250) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 1:10:45 loss 1.3971 (1.4619) acc 93.7500 (93.7500) lr 0.260000 +FPS@all 843.394, TIME@all 0.304 +epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:10:56 loss 1.5594 (1.4099) acc 90.6250 (94.8438) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:45 loss 1.4386 (1.4697) acc 93.7500 (93.6719) lr 0.260000 +FPS@all 843.358, TIME@all 0.304 +epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5365 (1.4119) acc 90.6250 (95.0000) lr 0.260000 +epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.4813 (1.4704) acc 87.5000 (93.2812) lr 0.260000 +FPS@all 843.189, TIME@all 0.304 +epoch: [71/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5959 (1.3871) acc 90.6250 (94.8438) lr 0.260000 +epoch: [71/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.5447 (1.4605) acc 87.5000 (93.5938) lr 0.260000 +FPS@all 843.544, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:10:36 loss 1.4566 (1.3945) acc 93.7500 (95.6250) lr 0.260000 +epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:10:25 loss 1.3345 (1.4249) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 843.810, TIME@all 0.303 +epoch: [72/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:10:36 loss 1.4857 (1.3731) acc 90.6250 (95.1562) lr 0.260000 +epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:10:25 loss 1.3887 (1.4192) acc 100.0000 (94.6094) lr 0.260000 +FPS@all 843.858, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:10:35 loss 1.5647 (1.4009) acc 90.6250 (94.8438) lr 0.260000 +epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:10:24 loss 1.3360 (1.4226) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 843.898, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:35 loss 1.3431 (1.3804) acc 96.8750 (96.4062) lr 0.260000 +epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:10:24 loss 1.5825 (1.4075) acc 90.6250 (95.5469) lr 0.260000 +FPS@all 843.936, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:35 loss 1.3913 (1.3710) acc 96.8750 (95.7812) lr 0.260000 +epoch: [72/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.3815 (1.4063) acc 96.8750 (95.0781) lr 0.260000 +FPS@all 844.207, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:10:35 loss 1.5704 (1.3970) acc 93.7500 (94.3750) lr 0.260000 +epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:10:24 loss 1.3494 (1.4203) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 844.029, TIME@all 0.303 +epoch: [72/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:10:36 loss 1.3494 (1.3796) acc 96.8750 (95.7812) lr 0.260000 +epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.5279 (1.4292) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 843.822, TIME@all 0.303 +epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:36 loss 1.4673 (1.4009) acc 93.7500 (95.7812) lr 0.260000 +epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.4637 (1.4466) acc 100.0000 (94.4531) lr 0.260000 +FPS@all 843.844, TIME@all 0.303 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5737 (1.3798) acc 90.6250 (95.0000) lr 0.260000 +epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.4560 (1.3907) acc 93.7500 (94.7656) lr 0.260000 +FPS@all 843.065, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:24 loss 1.6931 (1.3467) acc 93.7500 (96.2500) lr 0.260000 +epoch: [73/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.2662 (1.3898) acc 100.0000 (95.4688) lr 0.260000 +FPS@all 843.124, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:22 loss 1.3951 (1.3423) acc 93.7500 (96.8750) lr 0.260000 +epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:10:15 loss 1.3576 (1.3808) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 843.302, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:24 loss 1.4726 (1.3285) acc 93.7500 (97.0312) lr 0.260000 +epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.3291 (1.3577) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 843.072, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5604 (1.3383) acc 87.5000 (96.2500) lr 0.260000 +epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:16 loss 1.3358 (1.3643) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 843.134, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:22 loss 1.9414 (1.3586) acc 81.2500 (96.5625) lr 0.260000 +epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:10:16 loss 1.3189 (1.3808) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 843.216, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.4142 (1.3190) acc 96.8750 (97.3438) lr 0.260000 +epoch: [73/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:16 loss 1.3890 (1.3660) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 843.361, TIME@all 0.304 +epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5093 (1.3436) acc 87.5000 (97.0312) lr 0.260000 +epoch: [73/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.3533 (1.3593) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 843.103, TIME@all 0.304 +epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:09:57 loss 1.5269 (1.3388) acc 87.5000 (96.8750) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.4942 (1.3514) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 843.468, TIME@all 0.304 +epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:09:58 loss 1.3986 (1.3229) acc 93.7500 (96.5625) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.4687 (1.3485) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 843.495, TIME@all 0.303 +epoch: [74/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:09:57 loss 1.4649 (1.3232) acc 93.7500 (95.9375) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.6049 (1.3505) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 843.536, TIME@all 0.303 +epoch: [74/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 1:09:56 loss 1.8031 (1.3395) acc 87.5000 (96.7188) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:55 loss 1.5689 (1.3476) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 843.680, TIME@all 0.303 +epoch: [74/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:09:57 loss 1.3483 (1.3169) acc 96.8750 (97.0312) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.8912 (1.3613) acc 81.2500 (95.7812) lr 0.260000 +FPS@all 843.803, TIME@all 0.303 +epoch: [74/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:09:57 loss 1.6669 (1.3559) acc 90.6250 (97.0312) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.5122 (1.3670) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 843.450, TIME@all 0.304 +epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:09:58 loss 1.3065 (1.3187) acc 100.0000 (97.8125) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:09:56 loss 1.7678 (1.3600) acc 81.2500 (96.5625) lr 0.260000 +FPS@all 843.446, TIME@all 0.304 +epoch: [74/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 1:09:57 loss 1.3488 (1.3139) acc 96.8750 (97.3438) lr 0.260000 +epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.4525 (1.3542) acc 90.6250 (95.5469) lr 0.260000 +FPS@all 843.625, TIME@all 0.303 +epoch: [75/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:09:54 loss 1.6531 (1.3441) acc 90.6250 (96.8750) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.3539 (1.3823) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 843.044, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:53 loss 1.5347 (1.3426) acc 90.6250 (95.6250) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.4794 (1.3756) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 843.103, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:53 loss 1.3439 (1.3282) acc 96.8750 (96.8750) lr 0.260000 +epoch: [75/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:09:44 loss 1.5987 (1.3789) acc 87.5000 (95.3125) lr 0.260000 +FPS@all 843.109, TIME@all 0.304 +epoch: [75/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:09:54 loss 1.4990 (1.3361) acc 96.8750 (97.3438) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:45 loss 1.4720 (1.3733) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 843.052, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:09:53 loss 1.4803 (1.3099) acc 90.6250 (97.8125) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:44 loss 1.4857 (1.3691) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 843.236, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:54 loss 1.3383 (1.3448) acc 100.0000 (96.7188) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.3479 (1.3667) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 843.034, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:09:53 loss 1.4098 (1.3335) acc 87.5000 (97.0312) lr 0.260000 +epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:44 loss 1.3136 (1.3807) acc 96.8750 (95.3906) lr 0.260000 +FPS@all 843.184, TIME@all 0.304 +epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:52 loss 1.4255 (1.3198) acc 93.7500 (96.7188) lr 0.260000 +epoch: [75/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:09:44 loss 1.4892 (1.3904) acc 90.6250 (95.0000) lr 0.260000 +FPS@all 843.400, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:40 loss 1.2829 (1.2731) acc 100.0000 (99.0625) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:28 loss 1.3157 (1.3007) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.454, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:39 loss 1.3511 (1.2733) acc 93.7500 (98.4375) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.3461 (1.3067) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 843.469, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:09:39 loss 1.2740 (1.2828) acc 96.8750 (96.4062) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:09:28 loss 1.2339 (1.3072) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 843.395, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:40 loss 1.2665 (1.2598) acc 100.0000 (99.0625) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:28 loss 1.3039 (1.3007) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.434, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:39 loss 1.2532 (1.2859) acc 100.0000 (98.2812) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:09:28 loss 1.2610 (1.3036) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 843.438, TIME@all 0.304 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:09:38 loss 1.2981 (1.2839) acc 93.7500 (97.5000) lr 0.260000 +epoch: [76/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:09:27 loss 1.2194 (1.3139) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 843.635, TIME@all 0.303 +epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:09:39 loss 1.2465 (1.2814) acc 96.8750 (98.1250) lr 0.260000 +epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.2333 (1.3081) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 843.558, TIME@all 0.303 +epoch: [76/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:09:38 loss 1.3763 (1.3055) acc 93.7500 (97.5000) lr 0.260000 +epoch: [76/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.3706 (1.3125) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 843.800, TIME@all 0.303 +epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.3218 (1.2782) acc 93.7500 (97.3438) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.3280 (1.3258) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 843.916, TIME@all 0.303 +epoch: [77/350][20/50] time 0.300 (0.304) data 0.001 (0.014) eta 1:09:20 loss 1.3053 (1.2936) acc 100.0000 (97.1875) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4440 (1.3263) acc 90.6250 (96.1719) lr 0.260000 +FPS@all 843.960, TIME@all 0.303 +epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.2046 (1.2731) acc 100.0000 (98.7500) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4825 (1.3150) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 844.005, TIME@all 0.303 +epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.2135 (1.2899) acc 96.8750 (96.2500) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4221 (1.3293) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 843.946, TIME@all 0.303 +epoch: [77/350][20/50] time 0.299 (0.304) data 0.001 (0.014) eta 1:09:18 loss 1.3116 (1.2877) acc 96.8750 (98.1250) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:10 loss 1.3180 (1.3177) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 844.148, TIME@all 0.303 +epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 1:09:20 loss 1.2814 (1.2954) acc 96.8750 (98.1250) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.2931 (1.3301) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.931, TIME@all 0.303 +epoch: [77/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 1:09:19 loss 1.1955 (1.2923) acc 100.0000 (97.8125) lr 0.260000 +epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.2827 (1.3163) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 844.090, TIME@all 0.303 +epoch: [77/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 1:09:21 loss 1.2381 (1.2930) acc 100.0000 (97.3438) lr 0.260000 +epoch: [77/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:09:11 loss 1.3928 (1.3168) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 844.249, TIME@all 0.303 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:03 loss 1.3823 (1.2845) acc 93.7500 (96.7188) lr 0.260000 +epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:02 loss 1.3412 (1.3144) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 841.853, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:03 loss 1.3149 (1.2799) acc 93.7500 (97.6562) lr 0.260000 +epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2757 (1.3155) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 841.808, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:04 loss 1.3501 (1.3007) acc 90.6250 (97.8125) lr 0.260000 +epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2856 (1.3213) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 841.796, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:09:02 loss 1.3553 (1.2879) acc 96.8750 (97.5000) lr 0.260000 +epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:02 loss 1.4267 (1.3170) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 842.024, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.3148 (1.2841) acc 93.7500 (97.9688) lr 0.260000 +epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.3230 (1.3068) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 841.849, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.2928 (1.2980) acc 96.8750 (97.3438) lr 0.260000 +epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2648 (1.3272) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 841.823, TIME@all 0.304 +epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.3071 (1.2895) acc 96.8750 (97.5000) lr 0.260000 +epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:02 loss 1.4421 (1.3341) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 841.954, TIME@all 0.304 +epoch: [78/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:09:02 loss 1.3070 (1.2848) acc 93.7500 (97.3438) lr 0.260000 +epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.3565 (1.3178) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 842.176, TIME@all 0.304 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:08:47 loss 1.3355 (1.2888) acc 96.8750 (97.9688) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.4211 (1.3134) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 843.440, TIME@all 0.304 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:47 loss 1.2400 (1.2884) acc 96.8750 (97.1875) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.2535 (1.3106) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 843.435, TIME@all 0.304 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:46 loss 1.3063 (1.2949) acc 96.8750 (97.0312) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:41 loss 1.3666 (1.3190) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 843.506, TIME@all 0.303 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:47 loss 1.5164 (1.3154) acc 96.8750 (97.0312) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.4469 (1.3136) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 843.422, TIME@all 0.304 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:08:47 loss 1.4266 (1.3152) acc 93.7500 (96.5625) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:42 loss 1.3240 (1.3268) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 843.427, TIME@all 0.304 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:08:46 loss 1.3906 (1.2976) acc 96.8750 (98.1250) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:40 loss 1.2904 (1.3190) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 843.622, TIME@all 0.303 +epoch: [79/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:08:46 loss 1.4550 (1.2889) acc 90.6250 (97.5000) lr 0.260000 +epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:41 loss 1.3224 (1.3035) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 843.575, TIME@all 0.303 +epoch: [79/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:08:46 loss 1.3765 (1.2776) acc 96.8750 (97.9688) lr 0.260000 +epoch: [79/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:08:41 loss 1.5199 (1.3067) acc 87.5000 (97.1875) lr 0.260000 +FPS@all 843.832, TIME@all 0.303 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.3883 (1.3154) acc 96.8750 (97.0312) lr 0.260000 +epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:38 loss 1.2750 (1.3586) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 841.271, TIME@all 0.304 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.2339 (1.2560) acc 100.0000 (98.7500) lr 0.260000 +epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:39 loss 1.3034 (1.3180) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 841.222, TIME@all 0.304 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.2090 (1.3241) acc 100.0000 (96.8750) lr 0.260000 +epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:38 loss 1.3720 (1.3582) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 841.270, TIME@all 0.304 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.3533 (1.2988) acc 96.8750 (97.3438) lr 0.260000 +epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:08:39 loss 1.3261 (1.3431) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 841.258, TIME@all 0.304 +epoch: [80/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:08:46 loss 1.1980 (1.2917) acc 100.0000 (97.3438) lr 0.260000 +epoch: [80/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:08:39 loss 1.2387 (1.3384) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 841.239, TIME@all 0.304 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.1775 (1.2928) acc 100.0000 (98.1250) lr 0.260000 +epoch: [80/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 1:08:38 loss 1.2862 (1.3325) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 841.430, TIME@all 0.304 +epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.3085 (1.2993) acc 96.8750 (97.6562) lr 0.260000 +epoch: [80/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 1:08:38 loss 1.3851 (1.3243) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 841.379, TIME@all 0.304 +epoch: [80/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:08:43 loss 1.4225 (1.3103) acc 96.8750 (96.8750) lr 0.260000 +epoch: [80/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:08:39 loss 1.2010 (1.3449) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 841.623, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:23 loss 1.3057 (1.2922) acc 96.8750 (97.5000) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2956 (1.3065) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 841.310, TIME@all 0.304 +epoch: [81/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.3001 (1.2955) acc 100.0000 (97.9688) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.3819 (1.3212) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 841.351, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.2767 (1.2857) acc 100.0000 (98.1250) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2678 (1.3229) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 841.396, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:23 loss 1.4151 (1.3208) acc 93.7500 (97.8125) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.3037 (1.3250) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 841.343, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.2602 (1.3154) acc 100.0000 (96.7188) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2552 (1.3436) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 841.357, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:08:21 loss 1.4395 (1.3001) acc 96.8750 (97.6562) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:08:20 loss 1.2505 (1.2963) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 841.541, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:08:22 loss 1.2372 (1.3116) acc 96.8750 (97.3438) lr 0.260000 +epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:08:20 loss 1.2938 (1.3323) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 841.484, TIME@all 0.304 +epoch: [81/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:08:24 loss 1.3377 (1.3121) acc 100.0000 (97.5000) lr 0.260000 +epoch: [81/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:08:22 loss 1.3792 (1.3189) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 841.501, TIME@all 0.304 +epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:08:13 loss 1.3166 (1.3182) acc 100.0000 (97.0312) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:07:59 loss 1.3478 (1.3602) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 842.646, TIME@all 0.304 +epoch: [82/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.5052 (1.3446) acc 96.8750 (96.5625) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.3614 (1.3573) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 842.623, TIME@all 0.304 +epoch: [82/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.4148 (1.3096) acc 87.5000 (97.1875) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.3269 (1.3328) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.710, TIME@all 0.304 +epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:08:13 loss 1.3395 (1.3186) acc 100.0000 (96.5625) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.3500 (1.3602) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 842.649, TIME@all 0.304 +epoch: [82/350][20/50] time 0.308 (0.305) data 0.001 (0.013) eta 1:08:14 loss 1.4189 (1.3051) acc 96.8750 (97.1875) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:00 loss 1.4011 (1.3526) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 842.855, TIME@all 0.304 +epoch: [82/350][20/50] time 0.307 (0.305) data 0.001 (0.014) eta 1:08:12 loss 1.4017 (1.3010) acc 93.7500 (97.3438) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.5075 (1.3643) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 842.859, TIME@all 0.304 +epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.3687 (1.3137) acc 96.8750 (97.0312) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.2253 (1.3500) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 842.663, TIME@all 0.304 +epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:08:13 loss 1.5252 (1.3092) acc 96.8750 (97.8125) lr 0.260000 +epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.4037 (1.3336) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.766, TIME@all 0.304 +epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3078 (1.2577) acc 100.0000 (98.4375) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.3339 (1.3105) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.425, TIME@all 0.304 +epoch: [83/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:56 loss 1.2924 (1.2884) acc 96.8750 (98.1250) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:07:43 loss 1.1865 (1.3261) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 843.349, TIME@all 0.304 +epoch: [83/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:55 loss 1.2475 (1.2712) acc 100.0000 (98.4375) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:07:43 loss 1.3800 (1.3073) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 843.433, TIME@all 0.304 +epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:54 loss 1.3466 (1.2769) acc 93.7500 (97.9688) lr 0.260000 +epoch: [83/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:42 loss 1.2942 (1.3067) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 843.587, TIME@all 0.303 +epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.2896 (1.2674) acc 96.8750 (98.1250) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2962 (1.3212) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 843.384, TIME@all 0.304 +epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3808 (1.2604) acc 96.8750 (98.4375) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2489 (1.2942) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.387, TIME@all 0.304 +epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3395 (1.2778) acc 90.6250 (97.1875) lr 0.260000 +epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:42 loss 1.2456 (1.3142) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 843.519, TIME@all 0.303 +epoch: [83/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:07:54 loss 1.3370 (1.2709) acc 96.8750 (98.2812) lr 0.260000 +epoch: [83/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2786 (1.2993) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.667, TIME@all 0.303 +epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.2512 (1.2788) acc 100.0000 (97.8125) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:29 loss 1.4104 (1.3038) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.489, TIME@all 0.304 +epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.5222 (1.3292) acc 93.7500 (96.2500) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:28 loss 1.2064 (1.3408) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 842.538, TIME@all 0.304 +epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.2877 (1.3002) acc 100.0000 (97.5000) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:29 loss 1.2356 (1.3220) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 842.476, TIME@all 0.304 +epoch: [84/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:07:33 loss 1.3719 (1.3026) acc 96.8750 (97.3438) lr 0.260000 +epoch: [84/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:07:29 loss 1.3359 (1.3285) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 842.514, TIME@all 0.304 +epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:07:32 loss 1.4282 (1.3030) acc 96.8750 (97.6562) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:28 loss 1.2933 (1.3273) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 842.702, TIME@all 0.304 +epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:07:33 loss 1.2960 (1.2757) acc 100.0000 (98.2812) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:07:29 loss 1.3701 (1.3068) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 842.499, TIME@all 0.304 +epoch: [84/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:07:32 loss 1.2809 (1.2786) acc 100.0000 (98.2812) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:28 loss 1.2738 (1.3098) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 842.631, TIME@all 0.304 +epoch: [84/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:07:31 loss 1.2653 (1.2851) acc 100.0000 (97.3438) lr 0.260000 +epoch: [84/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:07:28 loss 1.2460 (1.3083) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 842.890, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:07:24 loss 1.5043 (1.3019) acc 84.3750 (96.5625) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.4409 (1.3296) acc 87.5000 (96.0938) lr 0.260000 +FPS@all 842.096, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:07:24 loss 1.4813 (1.3283) acc 93.7500 (96.4062) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.3964 (1.3490) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 842.033, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:07:24 loss 1.5987 (1.3423) acc 87.5000 (96.4062) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.4993 (1.3449) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 842.128, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.3742 (1.3277) acc 90.6250 (97.1875) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:07:18 loss 1.3074 (1.3383) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 842.216, TIME@all 0.304 +epoch: [85/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.4078 (1.3041) acc 93.7500 (97.0312) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:07:18 loss 1.4362 (1.3440) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 842.058, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.6190 (1.3200) acc 90.6250 (97.0312) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.3078 (1.3520) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 842.072, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.5979 (1.3283) acc 90.6250 (96.2500) lr 0.260000 +epoch: [85/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:07:17 loss 1.4457 (1.3576) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 842.423, TIME@all 0.304 +epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:07:23 loss 1.5305 (1.3236) acc 87.5000 (96.4062) lr 0.260000 +epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.007) eta 1:07:17 loss 1.5103 (1.3593) acc 84.3750 (95.5469) lr 0.260000 +FPS@all 842.269, TIME@all 0.304 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.3680 (1.3006) acc 93.7500 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.4668 (1.3404) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 843.962, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:58 loss 1.2946 (1.3033) acc 96.8750 (97.0312) lr 0.260000 +epoch: [86/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3861 (1.3354) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 843.908, TIME@all 0.303 +epoch: [86/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 1:06:56 loss 1.5178 (1.3082) acc 93.7500 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:06:53 loss 1.2489 (1.3342) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 844.034, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.6624 (1.3236) acc 87.5000 (96.2500) lr 0.260000 +epoch: [86/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3353 (1.3395) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 843.975, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:06:56 loss 1.4601 (1.3011) acc 84.3750 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:06:52 loss 1.4721 (1.3435) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 844.144, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.6571 (1.3200) acc 90.6250 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3458 (1.3413) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 843.964, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:56 loss 1.5813 (1.3161) acc 90.6250 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:52 loss 1.5208 (1.3536) acc 90.6250 (95.9375) lr 0.260000 +FPS@all 844.101, TIME@all 0.303 +epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:56 loss 1.5266 (1.3019) acc 93.7500 (97.5000) lr 0.260000 +epoch: [86/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:52 loss 1.5565 (1.3394) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 844.327, TIME@all 0.303 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:07:03 loss 1.2506 (1.2908) acc 100.0000 (97.9688) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.006) eta 1:06:53 loss 1.2087 (1.3390) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 841.067, TIME@all 0.304 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:03 loss 1.2664 (1.3017) acc 100.0000 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.3215 (1.3446) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 841.082, TIME@all 0.304 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.3143 (1.3196) acc 96.8750 (96.7188) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2395 (1.3516) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 841.096, TIME@all 0.304 +epoch: [87/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2574 (1.3028) acc 96.8750 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.3102 (1.3588) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 841.075, TIME@all 0.304 +epoch: [87/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:07:02 loss 1.2778 (1.3134) acc 96.8750 (97.9688) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.3752 (1.3627) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 841.222, TIME@all 0.304 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2818 (1.3144) acc 96.8750 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.2411 (1.3468) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 841.084, TIME@all 0.304 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:01 loss 1.4030 (1.3384) acc 93.7500 (96.2500) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2569 (1.3605) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 841.271, TIME@all 0.304 +epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2732 (1.2884) acc 100.0000 (97.8125) lr 0.260000 +epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2440 (1.3239) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 841.492, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.4015 (1.3529) acc 93.7500 (95.7812) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.5114 (1.3771) acc 96.8750 (95.0000) lr 0.260000 +FPS@all 842.459, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:39 loss 1.5629 (1.3676) acc 90.6250 (96.2500) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:30 loss 1.4610 (1.3882) acc 90.6250 (95.6250) lr 0.260000 +FPS@all 842.455, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.001 (0.013) eta 1:06:38 loss 1.3680 (1.3105) acc 93.7500 (97.6562) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:28 loss 1.3638 (1.3553) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 842.674, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:39 loss 1.2311 (1.3024) acc 96.8750 (96.7188) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:29 loss 1.3175 (1.3453) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 842.450, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:06:39 loss 1.3159 (1.3216) acc 96.8750 (96.8750) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.5484 (1.3424) acc 90.6250 (96.0938) lr 0.260000 +FPS@all 842.485, TIME@all 0.304 +epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:38 loss 1.4833 (1.3210) acc 87.5000 (96.8750) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:29 loss 1.3982 (1.3515) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 842.614, TIME@all 0.304 +epoch: [88/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.3303 (1.2922) acc 96.8750 (98.2812) lr 0.260000 +epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.3706 (1.3449) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 842.516, TIME@all 0.304 +epoch: [88/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.3866 (1.2928) acc 93.7500 (96.8750) lr 0.260000 +epoch: [88/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.4547 (1.3428) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 842.826, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 1:06:21 loss 1.3240 (1.2887) acc 93.7500 (97.8125) lr 0.260000 +epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.4713 (1.3166) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 842.486, TIME@all 0.304 +epoch: [89/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.2727 (1.2728) acc 100.0000 (98.5938) lr 0.260000 +epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.3718 (1.3193) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 842.559, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.3062 (1.2529) acc 96.8750 (99.2188) lr 0.260000 +epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.4767 (1.2991) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 842.507, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.3359 (1.2681) acc 96.8750 (97.6562) lr 0.260000 +epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.3399 (1.3054) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 842.499, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:06:20 loss 1.2950 (1.2754) acc 93.7500 (98.2812) lr 0.260000 +epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:12 loss 1.3956 (1.3040) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 842.717, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.6442 (1.2824) acc 87.5000 (98.4375) lr 0.260000 +epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.6008 (1.3052) acc 90.6250 (97.6562) lr 0.260000 +FPS@all 842.522, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.3531 (1.2718) acc 100.0000 (98.4375) lr 0.260000 +epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:12 loss 1.3198 (1.3012) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 842.668, TIME@all 0.304 +epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.1990 (1.2923) acc 100.0000 (97.8125) lr 0.260000 +epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:12 loss 1.4604 (1.3117) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 842.934, TIME@all 0.304 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:15 loss 1.2458 (1.3240) acc 96.8750 (97.5000) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.4084 (1.3597) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 840.598, TIME@all 0.305 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:15 loss 1.4187 (1.3121) acc 93.7500 (97.0312) lr 0.260000 +epoch: [90/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3094 (1.3460) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 840.623, TIME@all 0.305 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:14 loss 1.3199 (1.3259) acc 96.8750 (96.4062) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3052 (1.3503) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 840.682, TIME@all 0.305 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 1:06:13 loss 1.2030 (1.3251) acc 100.0000 (97.1875) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:06:10 loss 1.2912 (1.3439) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 840.837, TIME@all 0.304 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.3056 (1.3133) acc 93.7500 (96.5625) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3316 (1.3708) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 840.584, TIME@all 0.305 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.3841 (1.3494) acc 96.8750 (95.9375) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:10 loss 1.2966 (1.3570) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 840.786, TIME@all 0.304 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.2325 (1.3545) acc 100.0000 (96.2500) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.001 (0.006) eta 1:06:11 loss 1.2097 (1.3567) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 840.635, TIME@all 0.305 +epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:13 loss 1.3398 (1.3246) acc 96.8750 (96.8750) lr 0.260000 +epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:10 loss 1.2117 (1.3432) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 841.019, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.4597 (1.2901) acc 93.7500 (97.5000) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.5522 (1.3150) acc 87.5000 (96.7969) lr 0.260000 +FPS@all 842.822, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:50 loss 1.3264 (1.3039) acc 96.8750 (97.9688) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:42 loss 1.5275 (1.3374) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 842.977, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:51 loss 1.4144 (1.2841) acc 96.8750 (97.6562) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:05:43 loss 1.4038 (1.3115) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 842.840, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.3591 (1.2900) acc 96.8750 (97.5000) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.4482 (1.3142) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.857, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.5637 (1.3012) acc 93.7500 (97.1875) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3963 (1.3089) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 842.886, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:50 loss 1.5108 (1.3037) acc 84.3750 (97.0312) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:05:42 loss 1.3179 (1.3106) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 843.016, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.3534 (1.2847) acc 96.8750 (98.2812) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3823 (1.2977) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.819, TIME@all 0.304 +epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:50 loss 1.3278 (1.2711) acc 96.8750 (97.5000) lr 0.260000 +epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3288 (1.3014) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 843.178, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:44 loss 1.2489 (1.2452) acc 100.0000 (98.4375) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4842 (1.3022) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 841.514, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.3949 (1.2695) acc 93.7500 (97.9688) lr 0.260000 +epoch: [92/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:05:34 loss 1.4699 (1.3021) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 841.573, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.3386 (1.2521) acc 93.7500 (97.9688) lr 0.260000 +epoch: [92/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4060 (1.2868) acc 93.7500 (97.9688) lr 0.260000 +FPS@all 841.584, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:44 loss 1.2568 (1.2437) acc 96.8750 (98.4375) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:33 loss 1.2903 (1.2765) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 841.528, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:43 loss 1.2163 (1.2608) acc 100.0000 (97.8125) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:32 loss 1.2906 (1.2825) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 841.721, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:44 loss 1.3547 (1.2651) acc 96.8750 (98.1250) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4468 (1.2912) acc 90.6250 (97.5781) lr 0.260000 +FPS@all 841.537, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.2545 (1.2508) acc 100.0000 (97.9688) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.001 (0.006) eta 1:05:31 loss 1.3204 (1.2920) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.029, TIME@all 0.304 +epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:43 loss 1.3453 (1.2549) acc 93.7500 (98.2812) lr 0.260000 +epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:32 loss 1.2809 (1.2774) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 841.685, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:21 loss 1.4016 (1.2908) acc 90.6250 (97.3438) lr 0.260000 +epoch: [93/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3133 (1.2927) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 842.492, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:20 loss 1.3386 (1.3010) acc 96.8750 (97.5000) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3373 (1.3061) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.529, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:21 loss 1.2309 (1.2839) acc 96.8750 (97.8125) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3185 (1.3058) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 842.415, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:20 loss 1.2818 (1.2688) acc 96.8750 (97.8125) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:12 loss 1.4917 (1.3029) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 842.633, TIME@all 0.304 +epoch: [93/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.2983 (1.2861) acc 93.7500 (97.6562) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:13 loss 1.3063 (1.3116) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.464, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:20 loss 1.2772 (1.3016) acc 100.0000 (97.0312) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:12 loss 1.1988 (1.3352) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 842.603, TIME@all 0.304 +epoch: [93/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.4949 (1.3012) acc 93.7500 (97.6562) lr 0.260000 +epoch: [93/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:05:13 loss 1.3461 (1.3071) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.804, TIME@all 0.304 +epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.2957 (1.3020) acc 96.8750 (96.2500) lr 0.260000 +epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:13 loss 1.2735 (1.3346) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 842.451, TIME@all 0.304 +epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:05:05 loss 1.2798 (1.2789) acc 93.7500 (97.9688) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:05:00 loss 1.2124 (1.3031) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 841.874, TIME@all 0.304 +epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.2559 (1.2881) acc 100.0000 (97.6562) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.3963 (1.3114) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 841.880, TIME@all 0.304 +epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:05:05 loss 1.3142 (1.2739) acc 96.8750 (98.5938) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:05:00 loss 1.2559 (1.3048) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 841.936, TIME@all 0.304 +epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3647 (1.2883) acc 93.7500 (96.7188) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:05:00 loss 1.2402 (1.3037) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 841.886, TIME@all 0.304 +epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3048 (1.2751) acc 100.0000 (97.6562) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.4740 (1.2982) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 842.196, TIME@all 0.304 +epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:05:04 loss 1.2275 (1.2785) acc 100.0000 (98.1250) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:04:59 loss 1.2926 (1.3063) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 842.081, TIME@all 0.304 +epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3764 (1.2953) acc 93.7500 (97.0312) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.2897 (1.2976) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 841.894, TIME@all 0.304 +epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.2605 (1.2583) acc 100.0000 (97.9688) lr 0.260000 +epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:04:59 loss 1.4798 (1.3124) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 842.029, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2954 (1.2698) acc 96.8750 (97.9688) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 1:04:38 loss 1.3340 (1.3028) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 843.215, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2468 (1.2736) acc 100.0000 (98.7500) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:04:38 loss 1.3990 (1.3119) acc 93.7500 (97.8125) lr 0.260000 +FPS@all 843.181, TIME@all 0.304 +epoch: [95/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2942 (1.2780) acc 100.0000 (97.6562) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:04:38 loss 1.3878 (1.3065) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 843.266, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.2531 (1.2728) acc 100.0000 (97.9688) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.3919 (1.3215) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.215, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:43 loss 1.2275 (1.2497) acc 100.0000 (98.7500) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:37 loss 1.2781 (1.2791) acc 100.0000 (97.9688) lr 0.260000 +FPS@all 843.389, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.2736 (1.2613) acc 96.8750 (98.5938) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.2711 (1.2836) acc 96.8750 (98.2031) lr 0.260000 +FPS@all 843.219, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:43 loss 1.3501 (1.2765) acc 96.8750 (98.4375) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:37 loss 1.3989 (1.3184) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 843.350, TIME@all 0.304 +epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.4287 (1.3046) acc 93.7500 (97.1875) lr 0.260000 +epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.4178 (1.3230) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.638, TIME@all 0.303 +epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3958 (1.2964) acc 87.5000 (97.1875) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2558 (1.3348) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 842.938, TIME@all 0.304 +epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3680 (1.2969) acc 96.8750 (97.0312) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2060 (1.3043) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 842.983, TIME@all 0.304 +epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.4807 (1.2716) acc 93.7500 (97.3438) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2345 (1.3121) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 843.009, TIME@all 0.304 +epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:04:29 loss 1.2844 (1.3103) acc 100.0000 (97.1875) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:04:25 loss 1.1798 (1.3054) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 843.140, TIME@all 0.304 +epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.4624 (1.2954) acc 90.6250 (97.5000) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.3258 (1.3243) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 842.953, TIME@all 0.304 +epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:04:30 loss 1.4190 (1.3087) acc 96.8750 (97.6562) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:04:26 loss 1.2931 (1.3126) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.970, TIME@all 0.304 +epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3106 (1.2681) acc 100.0000 (97.5000) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.3530 (1.3053) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.241, TIME@all 0.304 +epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:04:29 loss 1.3375 (1.2693) acc 96.8750 (97.8125) lr 0.260000 +epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:04:25 loss 1.3358 (1.2889) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 843.114, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:04:30 loss 1.1695 (1.2635) acc 100.0000 (97.6562) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:04:21 loss 1.5265 (1.2895) acc 90.6250 (97.5000) lr 0.260000 +FPS@all 841.408, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:30 loss 1.2292 (1.2667) acc 96.8750 (98.4375) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.4445 (1.2901) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 841.432, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2617 (1.2498) acc 96.8750 (98.4375) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.2865 (1.2721) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 841.591, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:04:29 loss 1.2315 (1.2756) acc 100.0000 (97.6562) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:04:20 loss 1.3171 (1.2782) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 841.105, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2618 (1.2493) acc 100.0000 (98.9062) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.3217 (1.2797) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 841.551, TIME@all 0.304 +epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2591 (1.2625) acc 96.8750 (97.9688) lr 0.260000 +epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.4277 (1.3064) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 841.399, TIME@all 0.304 +epoch: [97/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:04:29 loss 1.1903 (1.2594) acc 100.0000 (97.8125) lr 0.260000 +epoch: [97/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:04:20 loss 1.3388 (1.2895) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 841.182, TIME@all 0.304 +epoch: [97/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:04:28 loss 1.2047 (1.2609) acc 100.0000 (98.1250) lr 0.260000 +epoch: [97/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.2923 (1.2992) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 841.670, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:03:55 loss 1.4851 (1.2749) acc 90.6250 (97.3438) lr 0.260000 +epoch: [98/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:58 loss 1.2640 (1.2886) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 842.343, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.3684 (1.2703) acc 93.7500 (98.2812) lr 0.260000 +epoch: [98/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:57 loss 1.4804 (1.2904) acc 90.6250 (97.5781) lr 0.260000 +FPS@all 842.362, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.4624 (1.2662) acc 93.7500 (98.1250) lr 0.260000 +epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.4303 (1.2901) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.409, TIME@all 0.304 +epoch: [98/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:03:55 loss 1.4632 (1.2617) acc 90.6250 (98.5938) lr 0.260000 +epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:58 loss 1.3854 (1.2998) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.339, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:55 loss 1.3994 (1.2891) acc 93.7500 (97.6562) lr 0.260000 +epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:58 loss 1.5809 (1.3074) acc 87.5000 (96.9531) lr 0.260000 +FPS@all 842.336, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:03:54 loss 1.3907 (1.2816) acc 93.7500 (97.0312) lr 0.260000 +epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.4892 (1.3030) acc 87.5000 (96.6406) lr 0.260000 +FPS@all 842.519, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.6576 (1.2833) acc 87.5000 (97.6562) lr 0.260000 +epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.2819 (1.2941) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 842.473, TIME@all 0.304 +epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.4326 (1.2737) acc 93.7500 (97.6562) lr 0.260000 +epoch: [98/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.5080 (1.2841) acc 90.6250 (97.5000) lr 0.260000 +FPS@all 842.709, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:54 loss 1.4885 (1.2922) acc 93.7500 (97.5000) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.3960 (1.3157) acc 90.6250 (96.9531) lr 0.260000 +FPS@all 842.556, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.3970 (1.2783) acc 96.8750 (97.3438) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2683 (1.3051) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 842.617, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.6742 (1.3117) acc 87.5000 (97.0312) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2714 (1.3211) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 842.532, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:53 loss 1.3677 (1.2969) acc 96.8750 (97.6562) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2363 (1.3082) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 842.552, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:52 loss 1.1993 (1.2978) acc 100.0000 (97.5000) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:41 loss 1.2642 (1.3220) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.756, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.3145 (1.2703) acc 93.7500 (97.9688) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.3393 (1.3030) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.552, TIME@all 0.304 +epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:53 loss 1.4020 (1.2863) acc 93.7500 (97.8125) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:41 loss 1.4293 (1.3184) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 842.711, TIME@all 0.304 +epoch: [99/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:03:50 loss 1.3998 (1.2831) acc 93.7500 (97.8125) lr 0.260000 +epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2675 (1.3141) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 842.932, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.2244 (1.2805) acc 100.0000 (97.8125) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:03:26 loss 1.2521 (1.3160) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 842.813, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.3322 (1.2754) acc 96.8750 (98.1250) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3548 (1.3122) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.887, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.4706 (1.2786) acc 93.7500 (98.2812) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.2693 (1.3024) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.835, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:03:33 loss 1.3026 (1.3010) acc 96.8750 (97.6562) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3877 (1.3196) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 842.826, TIME@all 0.304 +epoch: [100/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 1:03:32 loss 1.4139 (1.2917) acc 96.8750 (97.8125) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:25 loss 1.3959 (1.3114) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 843.031, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.3819 (1.2840) acc 96.8750 (98.1250) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.4425 (1.3198) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 842.815, TIME@all 0.304 +epoch: [100/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:03:32 loss 1.4303 (1.2862) acc 93.7500 (97.6562) lr 0.260000 +epoch: [100/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3051 (1.3041) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 843.178, TIME@all 0.304 +epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:03:33 loss 1.2208 (1.2729) acc 96.8750 (98.1250) lr 0.260000 +epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.4042 (1.3167) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 842.955, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.011) eta 1:03:17 loss 1.2758 (1.3116) acc 96.8750 (97.3438) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.3988 (1.3174) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 843.106, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2951 (1.3032) acc 96.8750 (97.5000) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2523 (1.3350) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 843.122, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.3241 (1.3297) acc 96.8750 (97.5000) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.3143 (1.3480) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 843.121, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 1:03:17 loss 1.2668 (1.3153) acc 100.0000 (96.8750) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:06 loss 1.2421 (1.3265) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.258, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 1:03:16 loss 1.4025 (1.3410) acc 87.5000 (96.7188) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:03:06 loss 1.4164 (1.3374) acc 90.6250 (96.7188) lr 0.260000 +FPS@all 843.316, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2200 (1.2974) acc 100.0000 (97.8125) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2886 (1.3198) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 843.114, TIME@all 0.304 +epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2578 (1.3023) acc 100.0000 (97.6562) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2949 (1.3351) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 843.109, TIME@all 0.304 +epoch: [101/350][20/50] time 0.309 (0.304) data 0.000 (0.012) eta 1:03:16 loss 1.2068 (1.3150) acc 100.0000 (96.8750) lr 0.260000 +epoch: [101/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:03:06 loss 1.2554 (1.3416) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 843.469, TIME@all 0.304 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:02:53 loss 1.3802 (1.3740) acc 93.7500 (96.0938) lr 0.260000 +epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:02:52 loss 1.3824 (1.3697) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 844.004, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:02:52 loss 1.3888 (1.3330) acc 96.8750 (96.5625) lr 0.260000 +epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3981 (1.3538) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 844.180, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.6200 (1.3466) acc 90.6250 (97.0312) lr 0.260000 +epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.3783 (1.3780) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 844.018, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.2498 (1.3330) acc 100.0000 (96.5625) lr 0.260000 +epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.3974 (1.3652) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 844.012, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:52 loss 1.3326 (1.3483) acc 96.8750 (96.8750) lr 0.260000 +epoch: [102/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3324 (1.3580) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 844.367, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:52 loss 1.2905 (1.3222) acc 100.0000 (97.3438) lr 0.260000 +epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.2915 (1.3672) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 844.072, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 1:02:52 loss 1.3555 (1.3190) acc 96.8750 (96.8750) lr 0.260000 +epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3751 (1.3427) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 844.215, TIME@all 0.303 +epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.4298 (1.3406) acc 93.7500 (96.5625) lr 0.260000 +epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.2613 (1.3588) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 844.045, TIME@all 0.303 +epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4078 (1.3129) acc 93.7500 (96.7188) lr 0.260000 +epoch: [103/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.3290 (1.3357) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 844.046, TIME@all 0.303 +epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:40 loss 1.3623 (1.2921) acc 93.7500 (97.8125) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:34 loss 1.2085 (1.3293) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 844.077, TIME@all 0.303 +epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4099 (1.3072) acc 93.7500 (95.7812) lr 0.260000 +epoch: [103/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.2867 (1.3121) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 844.148, TIME@all 0.303 +epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.2363 (1.2844) acc 100.0000 (97.1875) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:33 loss 1.3313 (1.3247) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 844.246, TIME@all 0.303 +epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.3069 (1.2800) acc 96.8750 (98.2812) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:34 loss 1.2012 (1.2993) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 844.065, TIME@all 0.303 +epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.3518 (1.3047) acc 96.8750 (97.9688) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 1:02:34 loss 1.3356 (1.3328) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 844.397, TIME@all 0.303 +epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.4207 (1.2950) acc 93.7500 (97.3438) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.2841 (1.3174) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 844.041, TIME@all 0.303 +epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4055 (1.3026) acc 96.8750 (97.6562) lr 0.260000 +epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.3186 (1.3310) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 844.181, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:24 loss 1.5898 (1.3247) acc 90.6250 (97.3438) lr 0.260000 +epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:02:17 loss 1.3666 (1.3456) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 844.117, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:23 loss 1.2613 (1.3049) acc 100.0000 (97.9688) lr 0.260000 +epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.3026 (1.3309) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 844.197, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:23 loss 1.2761 (1.3518) acc 100.0000 (96.5625) lr 0.260000 +epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.2783 (1.3441) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 844.158, TIME@all 0.303 +epoch: [104/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:24 loss 1.3617 (1.2978) acc 96.8750 (97.1875) lr 0.260000 +epoch: [104/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.3997 (1.3249) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 844.141, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:23 loss 1.4601 (1.3309) acc 93.7500 (96.8750) lr 0.260000 +epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.4657 (1.3466) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 844.145, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:23 loss 1.4244 (1.3290) acc 93.7500 (95.6250) lr 0.260000 +epoch: [104/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.2894 (1.3421) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 844.285, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:22 loss 1.2670 (1.3255) acc 93.7500 (96.5625) lr 0.260000 +epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.4765 (1.3560) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 844.345, TIME@all 0.303 +epoch: [104/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:02:23 loss 1.5102 (1.3221) acc 96.8750 (97.3438) lr 0.260000 +epoch: [104/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 1:02:16 loss 1.1758 (1.3470) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 844.504, TIME@all 0.303 +epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:08 loss 1.1906 (1.2827) acc 100.0000 (97.8125) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.5185 (1.3250) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 842.200, TIME@all 0.304 +epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.4717 (1.3153) acc 96.8750 (97.6562) lr 0.260000 +epoch: [105/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2903 (1.3278) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 842.275, TIME@all 0.304 +epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.2969 (1.3023) acc 96.8750 (97.3438) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2732 (1.3333) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 842.242, TIME@all 0.304 +epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:08 loss 1.3989 (1.3238) acc 93.7500 (96.5625) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2817 (1.3189) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.236, TIME@all 0.304 +epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:07 loss 1.3410 (1.3014) acc 96.8750 (97.6562) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:02:10 loss 1.2710 (1.3314) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 842.254, TIME@all 0.304 +epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.3283 (1.2951) acc 96.8750 (97.1875) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.4156 (1.3387) acc 90.6250 (95.9375) lr 0.260000 +FPS@all 842.556, TIME@all 0.304 +epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:02:07 loss 1.3207 (1.2740) acc 96.8750 (97.6562) lr 0.260000 +epoch: [105/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:02:09 loss 1.3241 (1.3176) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 842.410, TIME@all 0.304 +epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:07 loss 1.3284 (1.3071) acc 100.0000 (96.7188) lr 0.260000 +epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:02:09 loss 1.3385 (1.3271) acc 93.7500 (95.8594) lr 0.260000 +FPS@all 842.370, TIME@all 0.304 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.4231 (1.3414) acc 93.7500 (96.4062) lr 0.260000 +epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.3492 (1.3730) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 836.210, TIME@all 0.306 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.5058 (1.3506) acc 93.7500 (96.2500) lr 0.260000 +epoch: [106/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.4318 (1.3918) acc 90.6250 (95.1562) lr 0.260000 +FPS@all 836.191, TIME@all 0.306 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:26 loss 1.5742 (1.3312) acc 87.5000 (97.5000) lr 0.260000 +epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.4296 (1.3698) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 836.302, TIME@all 0.306 +epoch: [106/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.6270 (1.3435) acc 87.5000 (96.2500) lr 0.260000 +epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 1:02:19 loss 1.3431 (1.3529) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 836.419, TIME@all 0.306 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.4971 (1.3277) acc 93.7500 (96.8750) lr 0.260000 +epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.2449 (1.3682) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 836.220, TIME@all 0.306 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.013) eta 1:02:27 loss 1.5277 (1.3468) acc 93.7500 (97.3438) lr 0.260000 +epoch: [106/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 1:02:20 loss 1.3492 (1.3696) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 836.190, TIME@all 0.306 +epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.3071 (1.3750) acc 100.0000 (96.2500) lr 0.260000 +epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 1:02:19 loss 1.4583 (1.3877) acc 90.6250 (96.0938) lr 0.260000 +FPS@all 836.378, TIME@all 0.306 +epoch: [106/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.5378 (1.3741) acc 90.6250 (95.4688) lr 0.260000 +epoch: [106/350][40/50] time 0.310 (0.306) data 0.000 (0.007) eta 1:02:20 loss 1.4589 (1.3943) acc 90.6250 (95.0781) lr 0.260000 +FPS@all 836.564, TIME@all 0.306 +epoch: [107/350][20/50] time 0.304 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.3277 (1.2870) acc 96.8750 (97.6562) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2950 (1.3198) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 839.567, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.2343 (1.3079) acc 100.0000 (97.6562) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3900 (1.3457) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 839.612, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.5857 (1.3503) acc 90.6250 (96.5625) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.4229 (1.3519) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 839.553, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.2932 (1.3197) acc 100.0000 (97.0312) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3882 (1.3391) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 839.533, TIME@all 0.305 +epoch: [107/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.4367 (1.3276) acc 93.7500 (96.5625) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2554 (1.3504) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 839.933, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.001 (0.012) eta 1:02:03 loss 1.2361 (1.3216) acc 100.0000 (96.5625) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2967 (1.3589) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 839.539, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.3800 (1.3323) acc 96.8750 (95.6250) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3175 (1.3528) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 839.676, TIME@all 0.305 +epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:02:02 loss 1.2899 (1.3325) acc 96.8750 (96.5625) lr 0.260000 +epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3748 (1.3334) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 839.719, TIME@all 0.305 +epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.3595 (1.3169) acc 96.8750 (97.5000) lr 0.260000 +epoch: [108/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.4648 (1.3262) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 843.961, TIME@all 0.303 +epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:34 loss 1.3605 (1.3170) acc 100.0000 (97.1875) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.3866 (1.3475) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 844.003, TIME@all 0.303 +epoch: [108/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:01:33 loss 1.2888 (1.3212) acc 100.0000 (97.5000) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:01:19 loss 1.2815 (1.3437) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 843.985, TIME@all 0.303 +epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.2441 (1.2857) acc 100.0000 (97.6562) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.5555 (1.3102) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 843.990, TIME@all 0.303 +epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.4834 (1.3145) acc 96.8750 (97.1875) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.3798 (1.3401) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 844.036, TIME@all 0.303 +epoch: [108/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:33 loss 1.2918 (1.2976) acc 100.0000 (97.0312) lr 0.260000 +epoch: [108/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:01:18 loss 1.3992 (1.3233) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 844.128, TIME@all 0.303 +epoch: [108/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:32 loss 1.3835 (1.3312) acc 100.0000 (96.7188) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:01:18 loss 1.3855 (1.3421) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 844.170, TIME@all 0.303 +epoch: [108/350][20/50] time 0.309 (0.304) data 0.001 (0.013) eta 1:01:33 loss 1.3256 (1.3006) acc 93.7500 (97.1875) lr 0.260000 +epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:18 loss 1.3340 (1.3310) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 844.295, TIME@all 0.303 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:17 loss 1.2709 (1.2638) acc 100.0000 (98.2812) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:01:12 loss 1.1685 (1.2716) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 842.729, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:17 loss 1.2369 (1.2489) acc 100.0000 (98.2812) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:01:11 loss 1.2754 (1.2625) acc 96.8750 (97.8125) lr 0.260000 +FPS@all 842.817, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.2625 (1.2643) acc 100.0000 (98.2812) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.2364 (1.2655) acc 100.0000 (98.3594) lr 0.260000 +FPS@all 842.798, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:01:16 loss 1.2341 (1.2609) acc 100.0000 (98.4375) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:01:11 loss 1.3706 (1.2726) acc 96.8750 (98.1250) lr 0.260000 +FPS@all 842.968, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.3636 (1.2467) acc 96.8750 (98.9062) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:01:11 loss 1.2940 (1.2580) acc 93.7500 (98.4375) lr 0.260000 +FPS@all 842.755, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.2368 (1.2697) acc 100.0000 (98.2812) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:12 loss 1.2415 (1.2691) acc 100.0000 (98.2812) lr 0.260000 +FPS@all 842.770, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:01:17 loss 1.2131 (1.2544) acc 100.0000 (98.4375) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.4144 (1.2650) acc 87.5000 (97.9688) lr 0.260000 +FPS@all 842.895, TIME@all 0.304 +epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.1822 (1.2346) acc 100.0000 (98.9062) lr 0.260000 +epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.2715 (1.2609) acc 100.0000 (98.2031) lr 0.260000 +FPS@all 843.136, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:00 loss 1.3670 (1.2342) acc 96.8750 (98.5938) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.3073 (1.2839) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 841.269, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:00:59 loss 1.2480 (1.2901) acc 100.0000 (97.3438) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.1989 (1.3129) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 841.320, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:00:59 loss 1.3606 (1.2492) acc 93.7500 (98.2812) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.2446 (1.2872) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 841.294, TIME@all 0.304 +epoch: [110/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:00 loss 1.6144 (1.2714) acc 90.6250 (98.1250) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.3337 (1.3038) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 841.260, TIME@all 0.304 +epoch: [110/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:00:59 loss 1.2311 (1.2546) acc 100.0000 (98.9062) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:58 loss 1.3480 (1.3016) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 841.502, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:00:59 loss 1.2340 (1.2585) acc 100.0000 (98.1250) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.001 (0.006) eta 1:00:59 loss 1.2215 (1.2917) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 841.303, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:00 loss 1.1796 (1.2515) acc 100.0000 (97.9688) lr 0.260000 +epoch: [110/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:00:59 loss 1.2498 (1.2956) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 841.613, TIME@all 0.304 +epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:00:59 loss 1.4523 (1.2628) acc 93.7500 (97.8125) lr 0.260000 +epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:58 loss 1.2610 (1.2987) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 841.437, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2056 (1.2305) acc 100.0000 (98.5938) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:40 loss 1.4500 (1.2646) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 841.646, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2255 (1.2324) acc 100.0000 (98.7500) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:41 loss 1.2874 (1.2585) acc 93.7500 (98.1250) lr 0.260000 +FPS@all 841.539, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.1697 (1.2224) acc 100.0000 (99.0625) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:41 loss 1.2515 (1.2678) acc 100.0000 (97.9688) lr 0.260000 +FPS@all 841.592, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:49 loss 1.4148 (1.2345) acc 93.7500 (98.4375) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:00:40 loss 1.2550 (1.2774) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 841.583, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.2182 (1.2431) acc 96.8750 (98.7500) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:00:40 loss 1.1681 (1.2609) acc 100.0000 (98.4375) lr 0.260000 +FPS@all 841.754, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2645 (1.2582) acc 100.0000 (98.2812) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:00:40 loss 1.2473 (1.2818) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 841.585, TIME@all 0.304 +epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.3151 (1.2575) acc 100.0000 (98.2812) lr 0.260000 +epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:00:40 loss 1.3016 (1.2731) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 841.697, TIME@all 0.304 +epoch: [111/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.1676 (1.2415) acc 100.0000 (98.5938) lr 0.260000 +epoch: [111/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:00:40 loss 1.2874 (1.2728) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 841.930, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:00:45 loss 1.3093 (1.2650) acc 96.8750 (98.2812) lr 0.260000 +epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:00:30 loss 1.2168 (1.2925) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 842.171, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.3320 (1.2650) acc 93.7500 (97.8125) lr 0.260000 +epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.2311 (1.2960) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 842.265, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:46 loss 1.3578 (1.2740) acc 96.8750 (97.8125) lr 0.260000 +epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1968 (1.2935) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 842.218, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.2679 (1.2391) acc 100.0000 (98.9062) lr 0.260000 +epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:29 loss 1.1948 (1.2627) acc 100.0000 (97.9688) lr 0.260000 +FPS@all 842.342, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.4564 (1.2545) acc 96.8750 (97.8125) lr 0.260000 +epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:29 loss 1.1948 (1.3019) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 842.387, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.3382 (1.2392) acc 96.8750 (98.5938) lr 0.260000 +epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1653 (1.2728) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 842.157, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.001 (0.013) eta 1:00:45 loss 1.5002 (1.2702) acc 93.7500 (98.1250) lr 0.260000 +epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1963 (1.2911) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 842.193, TIME@all 0.304 +epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.4212 (1.2422) acc 93.7500 (98.7500) lr 0.260000 +epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1997 (1.2706) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 842.520, TIME@all 0.304 +epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:00:22 loss 1.2995 (1.2885) acc 100.0000 (97.8125) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:00:10 loss 1.3383 (1.2958) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 842.375, TIME@all 0.304 +epoch: [113/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2431 (1.3031) acc 100.0000 (96.4062) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:00:10 loss 1.3879 (1.3134) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 842.339, TIME@all 0.304 +epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2704 (1.2903) acc 100.0000 (97.9688) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.4671 (1.3211) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.406, TIME@all 0.304 +epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.3645 (1.2868) acc 90.6250 (97.3438) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.2594 (1.3028) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 842.377, TIME@all 0.304 +epoch: [113/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 1:00:21 loss 1.2271 (1.3004) acc 100.0000 (97.1875) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:09 loss 1.2839 (1.3111) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 842.575, TIME@all 0.304 +epoch: [113/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 1:00:22 loss 1.3390 (1.2867) acc 96.8750 (96.7188) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.2741 (1.3061) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 842.366, TIME@all 0.304 +epoch: [113/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 1:00:21 loss 1.2821 (1.2604) acc 96.8750 (98.2812) lr 0.260000 +epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:09 loss 1.2677 (1.2934) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 842.538, TIME@all 0.304 +epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2265 (1.2854) acc 100.0000 (97.1875) lr 0.260000 +epoch: [113/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.3106 (1.2944) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.702, TIME@all 0.304 +epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:55 loss 1.5116 (1.3313) acc 84.3750 (96.0938) lr 0.260000 +epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.3258 (1.3377) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 843.708, TIME@all 0.303 +epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:59:54 loss 1.5532 (1.3158) acc 90.6250 (96.4062) lr 0.260000 +epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.5179 (1.3413) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 843.774, TIME@all 0.303 +epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.4952 (1.3363) acc 93.7500 (96.2500) lr 0.260000 +epoch: [114/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.2965 (1.3583) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 843.743, TIME@all 0.303 +epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.5596 (1.3110) acc 87.5000 (97.0312) lr 0.260000 +epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:48 loss 1.3721 (1.3414) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 843.928, TIME@all 0.303 +epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:55 loss 1.4324 (1.3432) acc 90.6250 (97.0312) lr 0.260000 +epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.2720 (1.3354) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 843.737, TIME@all 0.303 +epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:54 loss 1.5527 (1.3340) acc 87.5000 (95.4688) lr 0.260000 +epoch: [114/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:59:48 loss 1.3685 (1.3320) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 843.786, TIME@all 0.303 +epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:59:53 loss 1.3550 (1.3133) acc 100.0000 (97.0312) lr 0.260000 +epoch: [114/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:48 loss 1.2578 (1.3259) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 844.078, TIME@all 0.303 +epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.6998 (1.3130) acc 90.6250 (96.8750) lr 0.260000 +epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:48 loss 1.2870 (1.3148) acc 90.6250 (97.1875) lr 0.260000 +FPS@all 843.871, TIME@all 0.303 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.2870 (1.2828) acc 96.8750 (97.9688) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.3474 (1.2936) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 842.980, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.2734 (1.2804) acc 100.0000 (97.9688) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.2895 (1.3160) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.008, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.3843 (1.2994) acc 93.7500 (97.0312) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:36 loss 1.3141 (1.3277) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 842.998, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:39 loss 1.3150 (1.2838) acc 96.8750 (97.6562) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:36 loss 1.3566 (1.3143) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.139, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.3794 (1.2930) acc 96.8750 (96.7188) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.3802 (1.3174) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 842.930, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:40 loss 1.3077 (1.2815) acc 96.8750 (98.4375) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:36 loss 1.2579 (1.3172) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.098, TIME@all 0.304 +epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:40 loss 1.3086 (1.2885) acc 100.0000 (97.8125) lr 0.260000 +epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:37 loss 1.4217 (1.2918) acc 90.6250 (97.4219) lr 0.260000 +FPS@all 842.950, TIME@all 0.304 +epoch: [115/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:59:39 loss 1.3145 (1.2737) acc 100.0000 (98.5938) lr 0.260000 +epoch: [115/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:36 loss 1.3237 (1.2974) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 843.272, TIME@all 0.304 +epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.3426 (1.2669) acc 96.8750 (98.1250) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.3648 (1.2888) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 843.837, TIME@all 0.303 +epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.2335 (1.2787) acc 100.0000 (97.9688) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.4575 (1.2954) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.937, TIME@all 0.303 +epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.1966 (1.2437) acc 100.0000 (98.5938) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:59:15 loss 1.5126 (1.2966) acc 90.6250 (97.6562) lr 0.260000 +FPS@all 843.903, TIME@all 0.303 +epoch: [116/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:59:20 loss 1.3063 (1.2766) acc 100.0000 (97.6562) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:59:15 loss 1.3393 (1.3039) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 843.903, TIME@all 0.303 +epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.2379 (1.2524) acc 100.0000 (97.9688) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.2761 (1.2916) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 843.883, TIME@all 0.303 +epoch: [116/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:59:19 loss 1.2715 (1.2512) acc 93.7500 (98.2812) lr 0.260000 +epoch: [116/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:14 loss 1.4063 (1.2831) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 844.083, TIME@all 0.303 +epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:59:20 loss 1.2319 (1.2940) acc 96.8750 (97.3438) lr 0.260000 +epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:59:14 loss 1.4744 (1.3153) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 844.022, TIME@all 0.303 +epoch: [116/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:59:22 loss 1.3257 (1.2621) acc 93.7500 (97.3438) lr 0.260000 +epoch: [116/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:59:16 loss 1.2083 (1.2863) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 844.081, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 0:59:03 loss 1.2283 (1.2584) acc 100.0000 (97.8125) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:59:01 loss 1.2560 (1.2959) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 843.835, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.4536 (1.2618) acc 93.7500 (98.7500) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:01 loss 1.1758 (1.2819) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 843.878, TIME@all 0.303 +epoch: [117/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:59:02 loss 1.3461 (1.2465) acc 93.7500 (97.9688) lr 0.260000 +epoch: [117/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:01 loss 1.2067 (1.2762) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 843.927, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.2918 (1.2771) acc 100.0000 (97.1875) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:59:01 loss 1.2850 (1.2953) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 843.855, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.3944 (1.2401) acc 93.7500 (98.7500) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:01 loss 1.2631 (1.2762) acc 100.0000 (98.0469) lr 0.260000 +FPS@all 843.845, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.001 (0.013) eta 0:59:02 loss 1.2862 (1.2453) acc 100.0000 (99.0625) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:00 loss 1.2983 (1.2888) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 844.064, TIME@all 0.303 +epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.4304 (1.2740) acc 90.6250 (97.0312) lr 0.260000 +epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:00 loss 1.2662 (1.2843) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 843.986, TIME@all 0.303 +epoch: [117/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:59:06 loss 1.3668 (1.2629) acc 96.8750 (98.5938) lr 0.260000 +epoch: [117/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:59:02 loss 1.4280 (1.2978) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 844.048, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.3826 (1.2686) acc 93.7500 (97.8125) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.4601 (1.3266) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 843.614, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2227 (1.2961) acc 96.8750 (96.7188) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.4673 (1.3248) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 843.602, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.4743 (1.3198) acc 90.6250 (96.5625) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.2148 (1.3401) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 843.541, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.4551 (1.2940) acc 96.8750 (97.6562) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3218 (1.3241) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 843.570, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:52 loss 1.2702 (1.2993) acc 100.0000 (97.1875) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:48 loss 1.2692 (1.3359) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 843.754, TIME@all 0.303 +epoch: [118/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2419 (1.3083) acc 100.0000 (97.3438) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3828 (1.3324) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 843.916, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2693 (1.2837) acc 96.8750 (97.9688) lr 0.260000 +epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3145 (1.3247) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 843.556, TIME@all 0.303 +epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:53 loss 1.3301 (1.3042) acc 100.0000 (96.8750) lr 0.260000 +epoch: [118/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:58:48 loss 1.4125 (1.3400) acc 90.6250 (96.0156) lr 0.260000 +FPS@all 843.701, TIME@all 0.303 +epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:45 loss 1.2189 (1.2832) acc 100.0000 (97.6562) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:58:40 loss 1.3084 (1.2955) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 842.016, TIME@all 0.304 +epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2108 (1.2569) acc 100.0000 (98.4375) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.4392 (1.2992) acc 90.6250 (97.3438) lr 0.260000 +FPS@all 842.037, TIME@all 0.304 +epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2346 (1.2726) acc 96.8750 (97.6562) lr 0.260000 +epoch: [119/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:39 loss 1.3725 (1.2920) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.089, TIME@all 0.304 +epoch: [119/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.3277 (1.2816) acc 96.8750 (98.2812) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.2696 (1.3094) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 842.034, TIME@all 0.304 +epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:45 loss 1.3439 (1.2869) acc 93.7500 (97.9688) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.3934 (1.3271) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.024, TIME@all 0.304 +epoch: [119/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.4097 (1.2543) acc 93.7500 (97.6562) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.3380 (1.2899) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.210, TIME@all 0.304 +epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.3419 (1.2608) acc 96.8750 (98.1250) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.2585 (1.2952) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 842.137, TIME@all 0.304 +epoch: [119/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2080 (1.2846) acc 100.0000 (97.1875) lr 0.260000 +epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.2368 (1.3090) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 842.457, TIME@all 0.304 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:58:16 loss 1.1914 (1.2792) acc 100.0000 (97.1875) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.2974 (1.3067) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.979, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:58:16 loss 1.2297 (1.2705) acc 100.0000 (98.2812) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.3054 (1.2884) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.930, TIME@all 0.303 +epoch: [120/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:58:17 loss 1.3647 (1.2727) acc 93.7500 (97.9688) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.2677 (1.2892) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 843.918, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.3146 (1.2765) acc 96.8750 (98.2812) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:15 loss 1.3980 (1.3016) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 843.959, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:58:16 loss 1.3402 (1.2999) acc 93.7500 (97.5000) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:14 loss 1.3401 (1.3204) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 844.136, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.2608 (1.2789) acc 100.0000 (97.6562) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:58:15 loss 1.3451 (1.3128) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 843.967, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.3470 (1.2971) acc 96.8750 (97.1875) lr 0.260000 +epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:14 loss 1.4234 (1.3148) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 844.109, TIME@all 0.303 +epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:15 loss 1.3630 (1.2733) acc 93.7500 (97.8125) lr 0.260000 +epoch: [120/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:15 loss 1.3325 (1.3007) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 844.355, TIME@all 0.303 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.3307 (1.2494) acc 96.8750 (97.9688) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2770 (1.2721) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 842.840, TIME@all 0.304 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.2181 (1.2451) acc 100.0000 (97.5000) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2270 (1.2738) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 842.857, TIME@all 0.304 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.3157 (1.2700) acc 96.8750 (97.5000) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2520 (1.3073) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 842.814, TIME@all 0.304 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:58:16 loss 1.2523 (1.2612) acc 100.0000 (98.1250) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2206 (1.2787) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.004, TIME@all 0.304 +epoch: [121/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.2706 (1.2674) acc 100.0000 (97.9688) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.3208 (1.2901) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 842.773, TIME@all 0.304 +epoch: [121/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.1784 (1.2333) acc 100.0000 (99.0625) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2866 (1.2683) acc 96.8750 (98.1250) lr 0.260000 +FPS@all 842.828, TIME@all 0.304 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:58:16 loss 1.1862 (1.2614) acc 100.0000 (97.8125) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2084 (1.2928) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 842.957, TIME@all 0.304 +epoch: [121/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:58:16 loss 1.2333 (1.2437) acc 100.0000 (98.9062) lr 0.260000 +epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.3476 (1.2845) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 843.176, TIME@all 0.304 +epoch: [122/350][20/50] time 0.305 (0.304) data 0.001 (0.012) eta 0:57:50 loss 1.3072 (1.2317) acc 100.0000 (98.4375) lr 0.260000 +epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.4415 (1.2919) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 843.283, TIME@all 0.304 +epoch: [122/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:57:51 loss 1.4054 (1.2612) acc 93.7500 (97.3438) lr 0.260000 +epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.3710 (1.2942) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 843.336, TIME@all 0.304 +epoch: [122/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.2822 (1.2384) acc 96.8750 (98.2812) lr 0.260000 +epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.2941 (1.2868) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 843.384, TIME@all 0.304 +epoch: [122/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:57:50 loss 1.2855 (1.2601) acc 96.8750 (97.9688) lr 0.260000 +epoch: [122/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:57:48 loss 1.4240 (1.2888) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 843.502, TIME@all 0.303 +epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:50 loss 1.3622 (1.2367) acc 96.8750 (98.5938) lr 0.260000 +epoch: [122/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:57:49 loss 1.3676 (1.2815) acc 93.7500 (97.8906) lr 0.260000 +FPS@all 843.313, TIME@all 0.304 +epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.3139 (1.2450) acc 93.7500 (98.2812) lr 0.260000 +epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.4563 (1.2813) acc 90.6250 (97.4219) lr 0.260000 +FPS@all 843.294, TIME@all 0.304 +epoch: [122/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:57:50 loss 1.2449 (1.2217) acc 96.8750 (98.9062) lr 0.260000 +epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:57:49 loss 1.5711 (1.2783) acc 90.6250 (97.3438) lr 0.260000 +FPS@all 843.453, TIME@all 0.304 +epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.2230 (1.2607) acc 100.0000 (98.5938) lr 0.260000 +epoch: [122/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.3406 (1.2784) acc 100.0000 (98.0469) lr 0.260000 +FPS@all 843.599, TIME@all 0.303 +epoch: [123/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:47 loss 1.2898 (1.2845) acc 100.0000 (97.6562) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.3849 (1.3257) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 842.197, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:47 loss 1.3526 (1.2825) acc 93.7500 (98.1250) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:40 loss 1.4097 (1.3212) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.214, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:57:46 loss 1.2299 (1.2789) acc 100.0000 (98.2812) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:39 loss 1.4965 (1.3326) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 842.277, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:47 loss 1.2677 (1.2923) acc 96.8750 (97.6562) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.4258 (1.3466) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 842.207, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:46 loss 1.2708 (1.2991) acc 96.8750 (97.5000) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:39 loss 1.4756 (1.3435) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 842.400, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:57:47 loss 1.2200 (1.2841) acc 100.0000 (97.8125) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:40 loss 1.4220 (1.3326) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 842.193, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:57:47 loss 1.1678 (1.2728) acc 100.0000 (97.9688) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.3882 (1.3073) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.481, TIME@all 0.304 +epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:46 loss 1.1637 (1.2799) acc 100.0000 (97.3438) lr 0.260000 +epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:39 loss 1.3679 (1.3214) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 842.339, TIME@all 0.304 +epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:32 loss 1.5447 (1.3529) acc 90.6250 (96.5625) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:57:23 loss 1.3545 (1.3664) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 841.915, TIME@all 0.304 +epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.4949 (1.3566) acc 93.7500 (95.7812) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:23 loss 1.2417 (1.3966) acc 100.0000 (95.3906) lr 0.260000 +FPS@all 841.960, TIME@all 0.304 +epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:32 loss 1.4692 (1.3580) acc 93.7500 (96.2500) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:57:23 loss 1.3137 (1.3787) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 841.925, TIME@all 0.304 +epoch: [124/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:57:33 loss 1.4357 (1.3764) acc 93.7500 (96.0938) lr 0.260000 +epoch: [124/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:57:23 loss 1.4052 (1.3976) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 841.990, TIME@all 0.304 +epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:57:31 loss 1.3650 (1.3395) acc 96.8750 (96.5625) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:22 loss 1.3434 (1.3789) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 842.144, TIME@all 0.304 +epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.7716 (1.3599) acc 87.5000 (96.5625) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:22 loss 1.2920 (1.3968) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 842.082, TIME@all 0.304 +epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:34 loss 1.4759 (1.3512) acc 100.0000 (96.7188) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:57:24 loss 1.2690 (1.3788) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 842.124, TIME@all 0.304 +epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.5004 (1.3540) acc 93.7500 (96.0938) lr 0.260000 +epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:23 loss 1.3394 (1.3831) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 841.939, TIME@all 0.304 +epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:10 loss 1.5197 (1.2731) acc 96.8750 (98.5938) lr 0.260000 +epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:57:07 loss 1.3869 (1.3123) acc 90.6250 (96.9531) lr 0.260000 +FPS@all 842.726, TIME@all 0.304 +epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:10 loss 1.3971 (1.2799) acc 93.7500 (97.6562) lr 0.260000 +epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:57:07 loss 1.2835 (1.3019) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 842.677, TIME@all 0.304 +epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.5004 (1.2926) acc 93.7500 (97.6562) lr 0.260000 +epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.4063 (1.3086) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.731, TIME@all 0.304 +epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:10 loss 1.5023 (1.3104) acc 90.6250 (97.0312) lr 0.260000 +epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.3727 (1.3137) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 842.692, TIME@all 0.304 +epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:57:09 loss 1.3305 (1.2718) acc 96.8750 (98.4375) lr 0.260000 +epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2658 (1.2836) acc 96.8750 (98.2031) lr 0.260000 +FPS@all 842.871, TIME@all 0.304 +epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:10 loss 1.2700 (1.2598) acc 96.8750 (98.5938) lr 0.260000 +epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.2065 (1.2850) acc 100.0000 (98.2031) lr 0.260000 +FPS@all 842.673, TIME@all 0.304 +epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.3521 (1.2769) acc 96.8750 (98.1250) lr 0.260000 +epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2957 (1.2999) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 842.829, TIME@all 0.304 +epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.2638 (1.2567) acc 100.0000 (98.5938) lr 0.260000 +epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2769 (1.2979) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.111, TIME@all 0.304 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:54 loss 1.3301 (1.2852) acc 100.0000 (97.1875) lr 0.260000 +epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.4258 (1.3075) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 843.675, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2665 (1.2653) acc 96.8750 (98.5938) lr 0.260000 +epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3919 (1.2846) acc 93.7500 (98.1250) lr 0.260000 +FPS@all 843.726, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2953 (1.2903) acc 100.0000 (96.8750) lr 0.260000 +epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3128 (1.2956) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 843.719, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2856 (1.2762) acc 96.8750 (97.5000) lr 0.260000 +epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.4890 (1.2903) acc 87.5000 (97.4219) lr 0.260000 +FPS@all 843.712, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:56:53 loss 1.2817 (1.2703) acc 100.0000 (97.9688) lr 0.260000 +epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:45 loss 1.2119 (1.2812) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.912, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3765 (1.2845) acc 100.0000 (97.3438) lr 0.260000 +epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:45 loss 1.3231 (1.2970) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 843.862, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3433 (1.2737) acc 93.7500 (98.1250) lr 0.260000 +epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3107 (1.2946) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.734, TIME@all 0.303 +epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3048 (1.2715) acc 100.0000 (97.8125) lr 0.260000 +epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.2328 (1.2991) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 843.990, TIME@all 0.303 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.2252 (1.2455) acc 100.0000 (98.4375) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.2098 (1.2662) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 842.702, TIME@all 0.304 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.4037 (1.2544) acc 96.8750 (98.1250) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3082 (1.2783) acc 96.8750 (98.1250) lr 0.260000 +FPS@all 842.764, TIME@all 0.304 +epoch: [127/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.2958 (1.2441) acc 96.8750 (98.2812) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3749 (1.2828) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.706, TIME@all 0.304 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.3346 (1.2550) acc 96.8750 (98.4375) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3842 (1.2832) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 842.728, TIME@all 0.304 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:56:38 loss 1.4448 (1.2742) acc 84.3750 (97.0312) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:56:36 loss 1.2842 (1.2813) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 842.899, TIME@all 0.304 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.2574 (1.2799) acc 100.0000 (97.9688) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.2237 (1.2901) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 842.704, TIME@all 0.304 +epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.3448 (1.2336) acc 93.7500 (98.2812) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:36 loss 1.3340 (1.2750) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.851, TIME@all 0.304 +epoch: [127/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.3450 (1.2406) acc 93.7500 (98.4375) lr 0.260000 +epoch: [127/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:56:36 loss 1.2658 (1.2842) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.127, TIME@all 0.304 +epoch: [128/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.3010 (1.2609) acc 96.8750 (98.1250) lr 0.260000 +epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.5283 (1.3024) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 843.855, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.2592 (1.2672) acc 96.8750 (97.9688) lr 0.260000 +epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.4982 (1.3007) acc 87.5000 (96.6406) lr 0.260000 +FPS@all 843.774, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:30 loss 1.4899 (1.2694) acc 93.7500 (97.5000) lr 0.260000 +epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.2180 (1.2923) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 843.837, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.1819 (1.2489) acc 100.0000 (98.2812) lr 0.260000 +epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.6716 (1.2875) acc 90.6250 (97.5000) lr 0.260000 +FPS@all 843.999, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.2873 (1.2548) acc 96.8750 (97.6562) lr 0.260000 +epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.3470 (1.2921) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 843.813, TIME@all 0.303 +epoch: [128/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2148 (1.2571) acc 96.8750 (97.9688) lr 0.260000 +epoch: [128/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.4824 (1.2898) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 844.056, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2150 (1.2338) acc 100.0000 (98.1250) lr 0.260000 +epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.3850 (1.2884) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 843.961, TIME@all 0.303 +epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2279 (1.2479) acc 100.0000 (98.7500) lr 0.260000 +epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.007) eta 0:56:18 loss 1.3729 (1.2891) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 843.795, TIME@all 0.303 +epoch: [129/350][20/50] time 0.307 (0.304) data 0.001 (0.014) eta 0:56:07 loss 1.3831 (1.3302) acc 93.7500 (96.5625) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.4259 (1.3338) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 843.472, TIME@all 0.304 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:07 loss 1.3206 (1.3045) acc 100.0000 (97.5000) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:56:00 loss 1.5771 (1.3303) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 843.477, TIME@all 0.304 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:07 loss 1.2487 (1.3046) acc 100.0000 (97.3438) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.3360 (1.3132) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 843.435, TIME@all 0.304 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:06 loss 1.2470 (1.3045) acc 96.8750 (97.5000) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.4581 (1.3265) acc 90.6250 (96.9531) lr 0.260000 +FPS@all 843.488, TIME@all 0.304 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:06 loss 1.3081 (1.3383) acc 93.7500 (96.4062) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:59 loss 1.3840 (1.3534) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 843.623, TIME@all 0.303 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:06 loss 1.4072 (1.3167) acc 90.6250 (97.0312) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.5290 (1.3467) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 843.442, TIME@all 0.304 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:06 loss 1.3488 (1.3219) acc 96.8750 (97.0312) lr 0.260000 +epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:59 loss 1.5394 (1.3534) acc 90.6250 (96.0938) lr 0.260000 +FPS@all 843.656, TIME@all 0.303 +epoch: [129/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:56:06 loss 1.2727 (1.3208) acc 100.0000 (97.5000) lr 0.260000 +epoch: [129/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.3048 (1.3335) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 843.810, TIME@all 0.303 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.011) eta 0:56:02 loss 1.3151 (1.3025) acc 96.8750 (97.6562) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:51 loss 1.4262 (1.3331) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 841.721, TIME@all 0.304 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:56:01 loss 1.3418 (1.3078) acc 96.8750 (96.7188) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:50 loss 1.2080 (1.3171) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 841.794, TIME@all 0.304 +epoch: [130/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:56:02 loss 1.3225 (1.3061) acc 96.8750 (97.0312) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:51 loss 1.2693 (1.3151) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 841.720, TIME@all 0.304 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2633 (1.3451) acc 96.8750 (96.2500) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:50 loss 1.3988 (1.3520) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 841.937, TIME@all 0.304 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2827 (1.2966) acc 96.8750 (97.8125) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:51 loss 1.4064 (1.3066) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 841.751, TIME@all 0.304 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.001 (0.012) eta 0:56:00 loss 1.2067 (1.2980) acc 100.0000 (97.0312) lr 0.260000 +epoch: [130/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:55:50 loss 1.3743 (1.3084) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 842.162, TIME@all 0.304 +epoch: [130/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.4782 (1.3249) acc 90.6250 (97.1875) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:50 loss 1.2729 (1.3214) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 841.880, TIME@all 0.304 +epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2626 (1.2688) acc 100.0000 (98.1250) lr 0.260000 +epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:51 loss 1.3367 (1.3020) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 841.733, TIME@all 0.304 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.011) eta 0:56:08 loss 1.3230 (1.2933) acc 93.7500 (96.5625) lr 0.260000 +epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3873 (1.3023) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 840.382, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:56:07 loss 1.4502 (1.2986) acc 96.8750 (97.3438) lr 0.260000 +epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.4301 (1.3253) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 840.480, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.001 (0.013) eta 0:56:07 loss 1.3671 (1.3041) acc 96.8750 (97.1875) lr 0.260000 +epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:55:45 loss 1.2571 (1.3261) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 840.622, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:56:08 loss 1.3987 (1.2724) acc 96.8750 (98.5938) lr 0.260000 +epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3235 (1.3125) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 840.440, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:56:08 loss 1.2898 (1.2953) acc 100.0000 (98.5938) lr 0.260000 +epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3192 (1.3035) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 840.417, TIME@all 0.305 +epoch: [131/350][20/50] time 0.303 (0.307) data 0.000 (0.012) eta 0:56:08 loss 1.4391 (1.2971) acc 93.7500 (97.6562) lr 0.260000 +epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.2889 (1.3130) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 840.668, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:56:07 loss 1.3273 (1.3153) acc 96.8750 (96.8750) lr 0.260000 +epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:45 loss 1.3891 (1.3282) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 840.573, TIME@all 0.305 +epoch: [131/350][20/50] time 0.304 (0.307) data 0.001 (0.012) eta 0:56:08 loss 1.3400 (1.2797) acc 93.7500 (97.8125) lr 0.260000 +epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.2739 (1.3050) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 840.436, TIME@all 0.305 +epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.5788 (1.2956) acc 93.7500 (97.6562) lr 0.260000 +epoch: [132/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.2804 (1.3407) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 841.942, TIME@all 0.304 +epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:55:28 loss 1.6019 (1.3219) acc 90.6250 (97.5000) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:55:20 loss 1.2238 (1.3444) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 841.844, TIME@all 0.304 +epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:55:27 loss 1.3400 (1.3129) acc 100.0000 (96.8750) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.1957 (1.3268) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 842.020, TIME@all 0.304 +epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.4667 (1.3281) acc 90.6250 (96.5625) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.3598 (1.3527) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 841.859, TIME@all 0.304 +epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:55:27 loss 1.5164 (1.3193) acc 87.5000 (96.7188) lr 0.260000 +epoch: [132/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.4052 (1.3712) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 842.062, TIME@all 0.304 +epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.4561 (1.3456) acc 93.7500 (96.7188) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.2597 (1.3521) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 841.855, TIME@all 0.304 +epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:28 loss 1.4913 (1.3118) acc 90.6250 (97.1875) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.2528 (1.3466) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 841.839, TIME@all 0.304 +epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:28 loss 1.4151 (1.3236) acc 96.8750 (96.5625) lr 0.260000 +epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.3242 (1.3492) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 842.163, TIME@all 0.304 +epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:55:12 loss 1.4685 (1.2909) acc 84.3750 (97.1875) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:55:07 loss 1.3594 (1.3224) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 842.208, TIME@all 0.304 +epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.2974 (1.2730) acc 96.8750 (97.9688) lr 0.260000 +epoch: [133/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3994 (1.3152) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.260, TIME@all 0.304 +epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.3856 (1.3053) acc 93.7500 (97.3438) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3608 (1.3235) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 842.248, TIME@all 0.304 +epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:11 loss 1.3169 (1.3368) acc 96.8750 (96.4062) lr 0.260000 +epoch: [133/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 0:55:07 loss 1.3026 (1.3335) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 842.232, TIME@all 0.304 +epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.4077 (1.3005) acc 93.7500 (96.5625) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.2074 (1.3139) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 842.231, TIME@all 0.304 +epoch: [133/350][20/50] time 0.309 (0.304) data 0.001 (0.014) eta 0:55:11 loss 1.2615 (1.2862) acc 96.8750 (96.8750) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:55:06 loss 1.2492 (1.3042) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 842.433, TIME@all 0.304 +epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:11 loss 1.2177 (1.2891) acc 100.0000 (97.9688) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:06 loss 1.4572 (1.3269) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.376, TIME@all 0.304 +epoch: [133/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:55:11 loss 1.3593 (1.3016) acc 90.6250 (97.3438) lr 0.260000 +epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3031 (1.3112) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 842.556, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:48 loss 1.3410 (1.2629) acc 96.8750 (98.7500) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.1839 (1.2956) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 843.122, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:54:47 loss 1.2577 (1.2678) acc 100.0000 (98.2812) lr 0.260000 +epoch: [134/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.2745 (1.2743) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 843.166, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:54:47 loss 1.2855 (1.2779) acc 96.8750 (98.2812) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.1966 (1.2881) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 843.072, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3113 (1.2574) acc 100.0000 (98.2812) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.4087 (1.2866) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.007, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:54:46 loss 1.2016 (1.2576) acc 100.0000 (98.2812) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:45 loss 1.2294 (1.2644) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 843.209, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3256 (1.2512) acc 96.8750 (97.9688) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.2105 (1.2693) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 843.004, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.4061 (1.2646) acc 96.8750 (98.7500) lr 0.260000 +epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:45 loss 1.2318 (1.2754) acc 100.0000 (98.2812) lr 0.260000 +FPS@all 843.236, TIME@all 0.304 +epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3654 (1.2820) acc 96.8750 (97.1875) lr 0.260000 +epoch: [134/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:54:46 loss 1.3079 (1.3012) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 843.417, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4973 (1.2762) acc 96.8750 (97.9688) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:33 loss 1.2814 (1.2954) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 842.593, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4306 (1.2951) acc 100.0000 (96.7188) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.2207 (1.2873) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 842.626, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.3652 (1.2919) acc 100.0000 (97.9688) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.2149 (1.3093) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.679, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.5105 (1.3126) acc 96.8750 (96.5625) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.3157 (1.3288) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 842.636, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:32 loss 1.2794 (1.2786) acc 100.0000 (97.8125) lr 0.260000 +epoch: [135/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:54:32 loss 1.3445 (1.2980) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 842.837, TIME@all 0.304 +epoch: [135/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:33 loss 1.4925 (1.3187) acc 87.5000 (97.5000) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:54:33 loss 1.2806 (1.3166) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 842.631, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4290 (1.3088) acc 96.8750 (96.7188) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.5640 (1.3185) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 842.989, TIME@all 0.304 +epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:33 loss 1.5069 (1.2775) acc 96.8750 (98.1250) lr 0.260000 +epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:54:32 loss 1.3361 (1.2906) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 842.779, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:28 loss 1.4885 (1.2563) acc 96.8750 (98.1250) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.3504 (1.2888) acc 90.6250 (96.7969) lr 0.260000 +FPS@all 843.019, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:28 loss 1.3968 (1.2489) acc 93.7500 (97.6562) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.2249 (1.2677) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 843.019, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:28 loss 1.2718 (1.2608) acc 100.0000 (98.7500) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:16 loss 1.2357 (1.2879) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 843.064, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:54:27 loss 1.2671 (1.2506) acc 100.0000 (98.7500) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:15 loss 1.2627 (1.2741) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.234, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:28 loss 1.3539 (1.2565) acc 93.7500 (97.9688) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:16 loss 1.2493 (1.2817) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 843.030, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:27 loss 1.3545 (1.2553) acc 96.8750 (97.9688) lr 0.260000 +epoch: [136/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.3209 (1.2908) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 843.095, TIME@all 0.304 +epoch: [136/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:27 loss 1.2738 (1.2745) acc 96.8750 (97.6562) lr 0.260000 +epoch: [136/350][40/50] time 0.298 (0.304) data 0.001 (0.007) eta 0:54:15 loss 1.3795 (1.2979) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 843.388, TIME@all 0.304 +epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:27 loss 1.3901 (1.2698) acc 87.5000 (97.5000) lr 0.260000 +epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:15 loss 1.3081 (1.2990) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 843.180, TIME@all 0.304 +epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:54:01 loss 1.3013 (1.2419) acc 93.7500 (97.6562) lr 0.260000 +epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:56 loss 1.3243 (1.2913) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 844.897, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:01 loss 1.3665 (1.2781) acc 93.7500 (97.6562) lr 0.260000 +epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:56 loss 1.4143 (1.3098) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 844.936, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:01 loss 1.1968 (1.2439) acc 96.8750 (98.2812) lr 0.260000 +epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:56 loss 1.2938 (1.2718) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 844.997, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:01 loss 1.3618 (1.2529) acc 96.8750 (98.4375) lr 0.260000 +epoch: [137/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:53:57 loss 1.3405 (1.2839) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 844.924, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:54:01 loss 1.3276 (1.2534) acc 96.8750 (98.4375) lr 0.260000 +epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:57 loss 1.2611 (1.2868) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 844.922, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:01 loss 1.3107 (1.2499) acc 96.8750 (98.1250) lr 0.260000 +epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:53:57 loss 1.2723 (1.2876) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 845.245, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:54:01 loss 1.2676 (1.2583) acc 100.0000 (97.5000) lr 0.260000 +epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:56 loss 1.4476 (1.2888) acc 84.3750 (97.1875) lr 0.260000 +FPS@all 845.105, TIME@all 0.303 +epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:00 loss 1.1975 (1.2634) acc 100.0000 (98.4375) lr 0.260000 +epoch: [137/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:53:56 loss 1.2043 (1.2909) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 845.084, TIME@all 0.303 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.4468 (1.3041) acc 93.7500 (98.1250) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.3801 (1.3294) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 842.757, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:53:52 loss 1.2151 (1.3108) acc 100.0000 (97.6562) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:53:48 loss 1.3431 (1.3306) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 842.706, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.2374 (1.3028) acc 100.0000 (96.4062) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.1926 (1.3151) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 842.731, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.2963 (1.3125) acc 96.8750 (96.7188) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.2661 (1.3374) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 842.748, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:53:51 loss 1.1919 (1.3064) acc 100.0000 (97.5000) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:47 loss 1.4500 (1.3444) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 842.954, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:50 loss 1.3882 (1.3031) acc 96.8750 (97.8125) lr 0.260000 +epoch: [138/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:53:47 loss 1.2686 (1.3283) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 842.960, TIME@all 0.304 +epoch: [138/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:53:51 loss 1.2739 (1.2964) acc 100.0000 (97.6562) lr 0.260000 +epoch: [138/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:53:47 loss 1.4284 (1.3238) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 843.145, TIME@all 0.304 +epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:53:52 loss 1.3631 (1.2925) acc 93.7500 (97.6562) lr 0.260000 +epoch: [138/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:48 loss 1.3716 (1.3240) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 842.883, TIME@all 0.304 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:53:32 loss 1.2804 (1.2813) acc 93.7500 (97.6562) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:28 loss 1.2219 (1.2864) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 844.142, TIME@all 0.303 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.2189 (1.2718) acc 100.0000 (97.9688) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:27 loss 1.4054 (1.3016) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 844.173, TIME@all 0.303 +epoch: [139/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.4167 (1.2809) acc 90.6250 (97.3438) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:28 loss 1.2979 (1.2921) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 844.186, TIME@all 0.303 +epoch: [139/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.3132 (1.2718) acc 93.7500 (98.2812) lr 0.260000 +epoch: [139/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:53:28 loss 1.3859 (1.2876) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 844.152, TIME@all 0.303 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:53:32 loss 1.3456 (1.2534) acc 100.0000 (98.2812) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.3660 (1.2698) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 844.359, TIME@all 0.303 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.2762 (1.2623) acc 96.8750 (98.2812) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:28 loss 1.3713 (1.2910) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 844.162, TIME@all 0.303 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.3446 (1.2559) acc 96.8750 (97.9688) lr 0.260000 +epoch: [139/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.5127 (1.2874) acc 87.5000 (97.4219) lr 0.260000 +FPS@all 844.550, TIME@all 0.303 +epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.4574 (1.2841) acc 93.7500 (97.0312) lr 0.260000 +epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.4456 (1.2921) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 844.311, TIME@all 0.303 +epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.011) eta 0:53:33 loss 1.3497 (1.2610) acc 96.8750 (98.1250) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.2169 (1.2794) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 842.093, TIME@all 0.304 +epoch: [140/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.3054 (1.2613) acc 100.0000 (98.5938) lr 0.260000 +epoch: [140/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 0:53:20 loss 1.2882 (1.2691) acc 96.8750 (98.2031) lr 0.260000 +FPS@all 842.211, TIME@all 0.304 +epoch: [140/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.4066 (1.2836) acc 96.8750 (98.1250) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.1786 (1.2829) acc 100.0000 (97.9688) lr 0.260000 +FPS@all 842.101, TIME@all 0.304 +epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:53:32 loss 1.2868 (1.2683) acc 100.0000 (98.2812) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:53:20 loss 1.2208 (1.2822) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 842.144, TIME@all 0.304 +epoch: [140/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.3662 (1.2902) acc 100.0000 (96.8750) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.3515 (1.2887) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 842.116, TIME@all 0.304 +epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:53:32 loss 1.3697 (1.2700) acc 93.7500 (97.5000) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:19 loss 1.2047 (1.2730) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 842.301, TIME@all 0.304 +epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:53:32 loss 1.3897 (1.2696) acc 96.8750 (98.1250) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:53:20 loss 1.1839 (1.2732) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 842.252, TIME@all 0.304 +epoch: [140/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:53:30 loss 1.3367 (1.2681) acc 96.8750 (98.9062) lr 0.260000 +epoch: [140/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:53:19 loss 1.3236 (1.2716) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 842.641, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:53:07 loss 1.3378 (1.2846) acc 100.0000 (97.8125) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:58 loss 1.7600 (1.3056) acc 84.3750 (97.0312) lr 0.260000 +FPS@all 843.200, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.3527 (1.2600) acc 96.8750 (97.8125) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4446 (1.3154) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 843.261, TIME@all 0.304 +epoch: [141/350][20/50] time 0.309 (0.304) data 0.000 (0.012) eta 0:53:06 loss 1.3650 (1.2747) acc 96.8750 (97.0312) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:58 loss 1.3453 (1.3018) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 843.275, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.4077 (1.2831) acc 93.7500 (98.2812) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4448 (1.3049) acc 90.6250 (97.3438) lr 0.260000 +FPS@all 843.199, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.3782 (1.2639) acc 93.7500 (97.9688) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4685 (1.2923) acc 90.6250 (97.5000) lr 0.260000 +FPS@all 843.192, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.2522 (1.2732) acc 96.8750 (97.9688) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:57 loss 1.3590 (1.3030) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.402, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.3373 (1.2694) acc 100.0000 (97.9688) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.5739 (1.3055) acc 87.5000 (96.7969) lr 0.260000 +FPS@all 843.350, TIME@all 0.304 +epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.2546 (1.2472) acc 100.0000 (97.9688) lr 0.260000 +epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.5873 (1.2970) acc 90.6250 (96.7969) lr 0.260000 +FPS@all 843.485, TIME@all 0.304 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:52:47 loss 1.3005 (1.3046) acc 96.8750 (97.0312) lr 0.260000 +epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:52:41 loss 1.3749 (1.2980) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 843.916, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:52:47 loss 1.3140 (1.2776) acc 90.6250 (97.1875) lr 0.260000 +epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:52:41 loss 1.2937 (1.2783) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 843.939, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2443 (1.2536) acc 96.8750 (98.2812) lr 0.260000 +epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.2564 (1.2779) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 843.957, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.4135 (1.2860) acc 96.8750 (97.8125) lr 0.260000 +epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.3137 (1.2883) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 843.935, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2621 (1.2668) acc 96.8750 (98.5938) lr 0.260000 +epoch: [142/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:40 loss 1.1941 (1.2912) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 844.106, TIME@all 0.303 +epoch: [142/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2294 (1.2779) acc 96.8750 (97.8125) lr 0.260000 +epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.3028 (1.2910) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 843.950, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2008 (1.2680) acc 100.0000 (97.5000) lr 0.260000 +epoch: [142/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.1670 (1.2790) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 844.089, TIME@all 0.303 +epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2659 (1.2670) acc 96.8750 (97.9688) lr 0.260000 +epoch: [142/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:52:41 loss 1.3597 (1.2864) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 844.297, TIME@all 0.303 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:52:38 loss 1.2547 (1.2643) acc 96.8750 (97.3438) lr 0.260000 +epoch: [143/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:31 loss 1.3084 (1.3038) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 842.927, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.2349 (1.2564) acc 96.8750 (97.8125) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2230 (1.2854) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 842.977, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2960 (1.2776) acc 96.8750 (97.6562) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2353 (1.2989) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 843.021, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.014) eta 0:52:37 loss 1.3392 (1.2860) acc 100.0000 (97.9688) lr 0.260000 +epoch: [143/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2458 (1.3078) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 843.143, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2509 (1.2689) acc 96.8750 (98.4375) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.3058 (1.2996) acc 90.6250 (97.4219) lr 0.260000 +FPS@all 842.945, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.3878 (1.2852) acc 93.7500 (97.1875) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2037 (1.2973) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 842.911, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2633 (1.2418) acc 100.0000 (98.1250) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2229 (1.2585) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 843.081, TIME@all 0.304 +epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.2147 (1.2524) acc 100.0000 (97.9688) lr 0.260000 +epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2126 (1.2982) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 843.311, TIME@all 0.304 +epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:52:26 loss 1.2046 (1.2459) acc 100.0000 (98.5938) lr 0.260000 +epoch: [144/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.3119 (1.2837) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.192, TIME@all 0.304 +epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:52:26 loss 1.2397 (1.2685) acc 100.0000 (97.6562) lr 0.260000 +epoch: [144/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.2591 (1.2780) acc 96.8750 (97.8125) lr 0.260000 +FPS@all 843.270, TIME@all 0.304 +epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.2295 (1.2388) acc 96.8750 (98.9062) lr 0.260000 +epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:15 loss 1.3125 (1.2844) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 843.238, TIME@all 0.304 +epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.1786 (1.2522) acc 100.0000 (98.1250) lr 0.260000 +epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:15 loss 1.3192 (1.2845) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 843.178, TIME@all 0.304 +epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:52:25 loss 1.3183 (1.2619) acc 96.8750 (97.8125) lr 0.260000 +epoch: [144/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:52:14 loss 1.3035 (1.3094) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 843.400, TIME@all 0.304 +epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.3332 (1.2713) acc 96.8750 (97.8125) lr 0.260000 +epoch: [144/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.2593 (1.2839) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 843.232, TIME@all 0.304 +epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:25 loss 1.1654 (1.2632) acc 100.0000 (97.5000) lr 0.260000 +epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:14 loss 1.2565 (1.2890) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 843.358, TIME@all 0.304 +epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.2032 (1.2430) acc 100.0000 (98.9062) lr 0.260000 +epoch: [144/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:14 loss 1.2541 (1.2743) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 843.560, TIME@all 0.303 +epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.4142 (1.3455) acc 96.8750 (95.6250) lr 0.260000 +epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4686 (1.3423) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 842.557, TIME@all 0.304 +epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:05 loss 1.5974 (1.3207) acc 93.7500 (96.7188) lr 0.260000 +epoch: [145/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4272 (1.3399) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 842.626, TIME@all 0.304 +epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:52:06 loss 1.3909 (1.3285) acc 93.7500 (96.8750) lr 0.260000 +epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:00 loss 1.5932 (1.3530) acc 87.5000 (96.2500) lr 0.260000 +FPS@all 842.600, TIME@all 0.304 +epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:52:05 loss 1.3390 (1.2996) acc 96.8750 (97.5000) lr 0.260000 +epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:59 loss 1.3974 (1.3342) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 842.772, TIME@all 0.304 +epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.4634 (1.3178) acc 96.8750 (97.3438) lr 0.260000 +epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.6142 (1.3392) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 842.603, TIME@all 0.304 +epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.3526 (1.3423) acc 96.8750 (95.3125) lr 0.260000 +epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4392 (1.3519) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 842.577, TIME@all 0.304 +epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:05 loss 1.4485 (1.3457) acc 96.8750 (96.5625) lr 0.260000 +epoch: [145/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.3340 (1.3589) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 842.898, TIME@all 0.304 +epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:52:05 loss 1.3380 (1.3118) acc 96.8750 (97.0312) lr 0.260000 +epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:51:59 loss 1.4189 (1.3398) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 842.722, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:47 loss 1.4851 (1.2992) acc 93.7500 (96.8750) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:48 loss 1.3972 (1.3363) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 842.487, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:46 loss 1.5751 (1.3134) acc 90.6250 (96.7188) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:47 loss 1.2241 (1.3421) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 842.574, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.6488 (1.2923) acc 87.5000 (97.1875) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.2787 (1.3196) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 842.504, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.3483 (1.2864) acc 90.6250 (97.3438) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.3993 (1.3147) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 842.519, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.4727 (1.2816) acc 93.7500 (97.6562) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.2159 (1.3247) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 842.520, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:46 loss 1.3292 (1.2780) acc 93.7500 (97.5000) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:47 loss 1.2315 (1.3380) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 842.667, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:51:46 loss 1.5755 (1.3187) acc 93.7500 (96.4062) lr 0.260000 +epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:47 loss 1.2692 (1.3715) acc 100.0000 (95.0781) lr 0.260000 +FPS@all 842.731, TIME@all 0.304 +epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.4629 (1.2978) acc 93.7500 (97.1875) lr 0.260000 +epoch: [146/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.4002 (1.3155) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 842.821, TIME@all 0.304 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.3242 (1.2781) acc 93.7500 (97.5000) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:51:29 loss 1.2298 (1.2905) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 843.550, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.2286 (1.2728) acc 100.0000 (97.3438) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:51:30 loss 1.3611 (1.3113) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 843.501, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.3143 (1.2755) acc 100.0000 (97.8125) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:51:29 loss 1.2684 (1.3022) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 843.605, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:38 loss 1.2725 (1.2707) acc 100.0000 (97.3438) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:29 loss 1.2980 (1.2998) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 843.650, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:51:38 loss 1.3720 (1.2705) acc 96.8750 (97.3438) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:51:30 loss 1.2759 (1.3047) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 843.532, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:39 loss 1.3223 (1.2836) acc 93.7500 (97.1875) lr 0.260000 +epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:30 loss 1.2614 (1.2944) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 843.536, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:38 loss 1.3368 (1.2621) acc 93.7500 (98.2812) lr 0.260000 +epoch: [147/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:51:29 loss 1.2227 (1.2987) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 843.939, TIME@all 0.303 +epoch: [147/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:51:38 loss 1.3327 (1.2642) acc 100.0000 (99.2188) lr 0.260000 +epoch: [147/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:51:29 loss 1.2200 (1.2868) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 843.694, TIME@all 0.303 +epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:20 loss 1.3826 (1.2724) acc 90.6250 (97.3438) lr 0.260000 +epoch: [148/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3571 (1.2875) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 844.688, TIME@all 0.303 +epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:20 loss 1.4701 (1.2619) acc 90.6250 (98.1250) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3022 (1.2906) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 844.643, TIME@all 0.303 +epoch: [148/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:19 loss 1.3454 (1.2777) acc 96.8750 (97.9688) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3518 (1.3155) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 844.714, TIME@all 0.303 +epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:19 loss 1.2432 (1.2407) acc 96.8750 (98.5938) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3764 (1.2729) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 844.842, TIME@all 0.303 +epoch: [148/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:20 loss 1.3008 (1.2571) acc 96.8750 (98.4375) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:09 loss 1.3254 (1.2804) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 844.642, TIME@all 0.303 +epoch: [148/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:51:20 loss 1.3215 (1.2474) acc 100.0000 (98.4375) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3757 (1.2817) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 844.646, TIME@all 0.303 +epoch: [148/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:51:19 loss 1.4479 (1.2805) acc 93.7500 (96.7188) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3789 (1.2902) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 845.045, TIME@all 0.303 +epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:19 loss 1.2521 (1.2636) acc 100.0000 (98.1250) lr 0.260000 +epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.2954 (1.2999) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 844.772, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:05 loss 1.2033 (1.2302) acc 100.0000 (98.4375) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:50:56 loss 1.3614 (1.2655) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 843.965, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:05 loss 1.2975 (1.2570) acc 96.8750 (97.9688) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:50:56 loss 1.3808 (1.2745) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 843.999, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.1885 (1.2205) acc 100.0000 (99.0625) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:50:56 loss 1.3008 (1.2652) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 844.078, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:51:04 loss 1.2248 (1.2544) acc 100.0000 (98.4375) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.2481 (1.2697) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 844.186, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.2117 (1.2629) acc 100.0000 (98.5938) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:50:56 loss 1.2864 (1.2939) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 844.005, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.2307 (1.2376) acc 96.8750 (98.2812) lr 0.260000 +epoch: [149/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3132 (1.2689) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 844.350, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:04 loss 1.3027 (1.2624) acc 96.8750 (98.2812) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3052 (1.2851) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 844.131, TIME@all 0.303 +epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.1968 (1.2651) acc 100.0000 (97.8125) lr 0.260000 +epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3394 (1.2827) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 843.985, TIME@all 0.303 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:50:55 loss 1.2840 (1.2479) acc 96.8750 (98.2812) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.3233 (1.2834) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 842.873, TIME@all 0.304 +epoch: [150/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:50:54 loss 1.2961 (1.2648) acc 96.8750 (98.2812) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.2387 (1.2994) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 842.949, TIME@all 0.304 +epoch: [150/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:50:55 loss 1.3931 (1.2728) acc 90.6250 (97.5000) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.4986 (1.3017) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 842.866, TIME@all 0.304 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:55 loss 1.2101 (1.2501) acc 100.0000 (98.5938) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:46 loss 1.3625 (1.2889) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 842.873, TIME@all 0.304 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:50:55 loss 1.5794 (1.2983) acc 90.6250 (97.3438) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 0:50:46 loss 1.2514 (1.3188) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 842.882, TIME@all 0.304 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:54 loss 1.2837 (1.2239) acc 96.8750 (98.5938) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:45 loss 1.1729 (1.2521) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 843.062, TIME@all 0.304 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:54 loss 1.4042 (1.2425) acc 93.7500 (98.4375) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:45 loss 1.3278 (1.2879) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 843.046, TIME@all 0.304 +epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:55 loss 1.3825 (1.2691) acc 96.8750 (97.5000) lr 0.260000 +epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.3972 (1.2942) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 843.157, TIME@all 0.304 +epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:50:42 loss 1.1923 (1.2009) acc 100.0000 (99.2188) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.2724 (1.2144) acc 96.8750 (98.5938) lr 0.026000 +FPS@all 840.796, TIME@all 0.304 +epoch: [151/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:50:42 loss 1.1842 (1.1968) acc 96.8750 (99.2188) lr 0.026000 +epoch: [151/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.2199 (1.1989) acc 96.8750 (98.9062) lr 0.026000 +FPS@all 840.840, TIME@all 0.304 +epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:50:42 loss 1.1701 (1.2133) acc 100.0000 (98.9062) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.1742 (1.2046) acc 96.8750 (98.7500) lr 0.026000 +FPS@all 840.865, TIME@all 0.304 +epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:42 loss 1.2024 (1.2002) acc 100.0000 (99.0625) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:50:35 loss 1.2133 (1.2159) acc 96.8750 (98.2812) lr 0.026000 +FPS@all 840.836, TIME@all 0.304 +epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.2158 (1.1949) acc 96.8750 (98.7500) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:50:35 loss 1.1654 (1.1991) acc 100.0000 (98.7500) lr 0.026000 +FPS@all 841.031, TIME@all 0.304 +epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:50:42 loss 1.1677 (1.1974) acc 100.0000 (98.4375) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:50:35 loss 1.1279 (1.2000) acc 100.0000 (98.5156) lr 0.026000 +FPS@all 840.802, TIME@all 0.304 +epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.1618 (1.1910) acc 100.0000 (99.5312) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:50:35 loss 1.2328 (1.2101) acc 96.8750 (98.7500) lr 0.026000 +FPS@all 840.992, TIME@all 0.304 +epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.1792 (1.1944) acc 100.0000 (99.0625) lr 0.026000 +epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.1189 (1.1909) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 841.250, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:50:23 loss 1.2449 (1.1516) acc 100.0000 (99.8438) lr 0.026000 +epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:50:16 loss 1.1544 (1.1637) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 843.377, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.1719 (1.1470) acc 100.0000 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1322 (1.1484) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.414, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:22 loss 1.3518 (1.1615) acc 96.8750 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1401 (1.1577) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.445, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:23 loss 1.1718 (1.1454) acc 100.0000 (99.5312) lr 0.026000 +epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1207 (1.1498) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 843.399, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.2056 (1.1524) acc 100.0000 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1298 (1.1566) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 843.394, TIME@all 0.304 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:22 loss 1.1467 (1.1524) acc 100.0000 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1692 (1.1561) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 843.590, TIME@all 0.303 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.1412 (1.1553) acc 100.0000 (99.3750) lr 0.026000 +epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1523 (1.1611) acc 100.0000 (99.1406) lr 0.026000 +FPS@all 843.683, TIME@all 0.303 +epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:22 loss 1.2479 (1.1536) acc 96.8750 (99.0625) lr 0.026000 +epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1481 (1.1493) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 843.552, TIME@all 0.303 +epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:49:56 loss 1.1679 (1.1261) acc 96.8750 (99.8438) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1455 (1.1398) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 844.693, TIME@all 0.303 +epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:49:55 loss 1.1036 (1.1223) acc 100.0000 (100.0000) lr 0.026000 +epoch: [153/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1069 (1.1369) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 844.713, TIME@all 0.303 +epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:49:55 loss 1.1114 (1.1344) acc 100.0000 (99.5312) lr 0.026000 +epoch: [153/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1620 (1.1454) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 844.740, TIME@all 0.303 +epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:49:55 loss 1.1030 (1.1306) acc 100.0000 (99.6875) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.1125 (1.1326) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 844.718, TIME@all 0.303 +epoch: [153/350][20/50] time 0.302 (0.303) data 0.001 (0.013) eta 0:49:55 loss 1.1335 (1.1231) acc 100.0000 (99.8438) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.1625 (1.1312) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 844.728, TIME@all 0.303 +epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:49:56 loss 1.2230 (1.1339) acc 96.8750 (99.5312) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.2110 (1.1499) acc 100.0000 (99.2188) lr 0.026000 +FPS@all 844.953, TIME@all 0.303 +epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:49:55 loss 1.1162 (1.1204) acc 100.0000 (99.8438) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:49:53 loss 1.1904 (1.1307) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 844.929, TIME@all 0.303 +epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:49:55 loss 1.1435 (1.1249) acc 100.0000 (100.0000) lr 0.026000 +epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:53 loss 1.1257 (1.1394) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 844.879, TIME@all 0.303 +epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:49:46 loss 1.1424 (1.1158) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:41 loss 1.1761 (1.1358) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 843.874, TIME@all 0.303 +epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:49:47 loss 1.1103 (1.1228) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:41 loss 1.1205 (1.1254) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.797, TIME@all 0.303 +epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:49:46 loss 1.1088 (1.1198) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:40 loss 1.1180 (1.1221) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.875, TIME@all 0.303 +epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:47 loss 1.1326 (1.1238) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:41 loss 1.1363 (1.1299) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.803, TIME@all 0.303 +epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1852 (1.1185) acc 96.8750 (99.5312) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.2492 (1.1303) acc 96.8750 (99.4531) lr 0.026000 +FPS@all 844.031, TIME@all 0.303 +epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1272 (1.1178) acc 100.0000 (100.0000) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:41 loss 1.2030 (1.1312) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 843.829, TIME@all 0.303 +epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:49:47 loss 1.0971 (1.1152) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.1150 (1.1282) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 844.179, TIME@all 0.303 +epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1072 (1.1160) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.1274 (1.1249) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.982, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:49:29 loss 1.1423 (1.1230) acc 100.0000 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:22 loss 1.1196 (1.1349) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 845.019, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.2167 (1.1197) acc 96.8750 (99.8438) lr 0.026000 +epoch: [155/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1264 (1.1328) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 845.064, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.1419 (1.1064) acc 100.0000 (100.0000) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1134 (1.1365) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 845.054, TIME@all 0.303 +epoch: [155/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:49:30 loss 1.1795 (1.1107) acc 100.0000 (100.0000) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1544 (1.1314) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 845.038, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:49:29 loss 1.2041 (1.1208) acc 93.7500 (99.5312) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:49:21 loss 1.1249 (1.1339) acc 100.0000 (99.1406) lr 0.026000 +FPS@all 845.209, TIME@all 0.303 +epoch: [155/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.1658 (1.1224) acc 100.0000 (99.5312) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.0982 (1.1326) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 845.339, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:30 loss 1.1683 (1.1199) acc 100.0000 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1818 (1.1326) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 844.994, TIME@all 0.303 +epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:49:29 loss 1.1710 (1.1260) acc 100.0000 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:49:22 loss 1.1338 (1.1392) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 845.143, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.1388 (1.1041) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.0903 (1.1148) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 844.811, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.0997 (1.1083) acc 100.0000 (99.6875) lr 0.026000 +epoch: [156/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1464 (1.1216) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 844.798, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:49:14 loss 1.0892 (1.1119) acc 100.0000 (99.8438) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:49:07 loss 1.1542 (1.1208) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 844.920, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.0902 (1.1108) acc 100.0000 (99.8438) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1338 (1.1193) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 844.798, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:49:14 loss 1.0917 (1.1034) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:49:06 loss 1.2658 (1.1229) acc 96.8750 (99.3750) lr 0.026000 +FPS@all 845.028, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.0883 (1.1031) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1403 (1.1231) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 844.816, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.1078 (1.1099) acc 100.0000 (99.6875) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1519 (1.1287) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 845.143, TIME@all 0.303 +epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.1398 (1.1063) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1048 (1.1210) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 844.966, TIME@all 0.303 +epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:49:15 loss 1.0952 (1.1021) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 0:49:01 loss 1.1013 (1.1104) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.042, TIME@all 0.304 +epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0910 (1.1019) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:49:01 loss 1.0947 (1.1149) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.100, TIME@all 0.304 +epoch: [157/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0975 (1.1028) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.1159 (1.1144) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 842.100, TIME@all 0.304 +epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0913 (1.1063) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:49:01 loss 1.0961 (1.1102) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.093, TIME@all 0.304 +epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:14 loss 1.0940 (1.0987) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.1129 (1.1132) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.239, TIME@all 0.304 +epoch: [157/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.1015 (1.1005) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.0950 (1.1039) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.355, TIME@all 0.304 +epoch: [157/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:49:15 loss 1.0975 (1.1017) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:49:01 loss 1.1030 (1.1161) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 842.116, TIME@all 0.304 +epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:49:14 loss 1.1272 (1.1040) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:00 loss 1.1044 (1.1132) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.227, TIME@all 0.304 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.2468 (1.1108) acc 96.8750 (99.8438) lr 0.026000 +epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1951 (1.1185) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.842, TIME@all 0.303 +epoch: [158/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.3219 (1.1146) acc 96.8750 (99.6875) lr 0.026000 +epoch: [158/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1032 (1.1164) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.939, TIME@all 0.303 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.2199 (1.1019) acc 100.0000 (100.0000) lr 0.026000 +epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1025 (1.1072) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.897, TIME@all 0.303 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.3578 (1.1123) acc 90.6250 (99.5312) lr 0.026000 +epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.0821 (1.1149) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.890, TIME@all 0.303 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:48:46 loss 1.1550 (1.1084) acc 100.0000 (99.8438) lr 0.026000 +epoch: [158/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:48:40 loss 1.0950 (1.1179) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 844.074, TIME@all 0.303 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.1667 (1.1056) acc 96.8750 (99.5312) lr 0.026000 +epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1141 (1.1199) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 843.866, TIME@all 0.303 +epoch: [158/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:48:47 loss 1.1677 (1.1059) acc 100.0000 (99.8438) lr 0.026000 +epoch: [158/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:48:40 loss 1.1022 (1.1131) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 844.217, TIME@all 0.303 +epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:48:46 loss 1.1440 (1.1087) acc 100.0000 (99.6875) lr 0.026000 +epoch: [158/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:48:40 loss 1.1037 (1.1139) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 844.026, TIME@all 0.303 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:48:34 loss 1.1167 (1.1005) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:48:27 loss 1.1698 (1.1063) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 841.592, TIME@all 0.304 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:34 loss 1.1097 (1.0974) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:27 loss 1.1405 (1.1057) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.621, TIME@all 0.304 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:33 loss 1.1292 (1.1000) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:26 loss 1.2004 (1.1066) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 841.640, TIME@all 0.304 +epoch: [159/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1180 (1.1029) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:27 loss 1.1126 (1.1047) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.589, TIME@all 0.304 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1537 (1.0994) acc 100.0000 (99.6875) lr 0.026000 +epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:26 loss 1.1439 (1.1058) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.806, TIME@all 0.304 +epoch: [159/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:48:33 loss 1.0820 (1.1003) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:26 loss 1.1452 (1.1067) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.079, TIME@all 0.304 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:34 loss 1.1184 (1.0980) acc 100.0000 (99.8438) lr 0.026000 +epoch: [159/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:48:27 loss 1.1392 (1.1083) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.594, TIME@all 0.304 +epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1169 (1.1005) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:26 loss 1.1118 (1.1088) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.763, TIME@all 0.304 +epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.0881 (1.0885) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0969 (1.1051) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.116, TIME@all 0.304 +epoch: [160/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.1250 (1.0910) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0953 (1.1030) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.136, TIME@all 0.304 +epoch: [160/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:48:20 loss 1.0853 (1.0937) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0844 (1.1121) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 841.280, TIME@all 0.304 +epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.0951 (1.0944) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0857 (1.1029) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.145, TIME@all 0.304 +epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.1011 (1.0987) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0823 (1.1053) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.125, TIME@all 0.304 +epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:48:19 loss 1.1079 (1.0981) acc 100.0000 (99.5312) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:48:18 loss 1.0864 (1.1092) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 841.305, TIME@all 0.304 +epoch: [160/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 0:48:20 loss 1.0964 (1.0913) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0816 (1.1050) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.360, TIME@all 0.304 +epoch: [160/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:48:20 loss 1.0947 (1.0909) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.1002 (1.1068) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.166, TIME@all 0.304 +epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 0:48:12 loss 1.0931 (1.0967) acc 100.0000 (100.0000) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0871 (1.1057) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.134, TIME@all 0.305 +epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.1237 (1.1011) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0957 (1.1076) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.144, TIME@all 0.305 +epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0829 (1.1026) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0899 (1.1100) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 839.167, TIME@all 0.305 +epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:48:12 loss 1.0779 (1.0963) acc 100.0000 (99.6875) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0982 (1.1095) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.127, TIME@all 0.305 +epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0956 (1.0948) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0746 (1.1049) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.120, TIME@all 0.305 +epoch: [161/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:48:11 loss 1.0978 (1.1007) acc 100.0000 (99.5312) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:48:10 loss 1.0876 (1.1070) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.316, TIME@all 0.305 +epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0844 (1.1007) acc 100.0000 (100.0000) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.1241 (1.1150) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.258, TIME@all 0.305 +epoch: [161/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:48:11 loss 1.1024 (1.0966) acc 100.0000 (100.0000) lr 0.026000 +epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0808 (1.1063) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.413, TIME@all 0.305 +epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:48:07 loss 1.1145 (1.0934) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1645 (1.1066) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.068, TIME@all 0.305 +epoch: [162/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1203 (1.0902) acc 100.0000 (100.0000) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1372 (1.1020) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.179, TIME@all 0.305 +epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1127 (1.1077) acc 100.0000 (99.3750) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1637 (1.1080) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.146, TIME@all 0.305 +epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.1217 (1.0922) acc 100.0000 (100.0000) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1429 (1.1078) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.074, TIME@all 0.305 +epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1982 (1.0945) acc 96.8750 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.001 (0.006) eta 0:47:55 loss 1.1126 (1.0952) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.106, TIME@all 0.305 +epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.2037 (1.1017) acc 100.0000 (99.6875) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1161 (1.1092) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.305, TIME@all 0.305 +epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.1093 (1.0876) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1119 (1.0999) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.247, TIME@all 0.305 +epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:48:04 loss 1.1124 (1.0939) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:47:54 loss 1.0978 (1.0979) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.619, TIME@all 0.305 +epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 0:47:46 loss 1.0867 (1.0918) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:38 loss 1.1471 (1.1028) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.011, TIME@all 0.305 +epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:46 loss 1.1066 (1.0977) acc 100.0000 (99.6875) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:38 loss 1.0729 (1.1040) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.014, TIME@all 0.305 +epoch: [163/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1279 (1.0992) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:38 loss 1.0827 (1.1032) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.059, TIME@all 0.305 +epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:45 loss 1.0851 (1.0906) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:37 loss 1.1376 (1.1007) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.247, TIME@all 0.305 +epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1130 (1.1116) acc 100.0000 (99.5312) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1071 (1.1154) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.068, TIME@all 0.305 +epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:46 loss 1.1322 (1.1159) acc 100.0000 (99.2188) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:37 loss 1.1002 (1.1169) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 840.172, TIME@all 0.305 +epoch: [163/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1530 (1.1080) acc 100.0000 (99.5312) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1007 (1.1175) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 840.004, TIME@all 0.305 +epoch: [163/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.0855 (1.0955) acc 100.0000 (99.8438) lr 0.026000 +epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1165 (1.1144) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 840.342, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1784 (1.0929) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1743 (1.1061) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 838.984, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.012) eta 0:47:38 loss 1.1773 (1.0879) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.315 (0.306) data 0.000 (0.006) eta 0:47:28 loss 1.0898 (1.1083) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 838.895, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.001 (0.013) eta 0:47:37 loss 1.0977 (1.0922) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1716 (1.1070) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.960, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.2041 (1.0983) acc 96.8750 (99.5312) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:27 loss 1.1205 (1.1057) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.072, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.2038 (1.1023) acc 96.8750 (99.6875) lr 0.026000 +epoch: [164/350][40/50] time 0.315 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1168 (1.1093) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 838.928, TIME@all 0.305 +epoch: [164/350][20/50] time 0.311 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1105 (1.0909) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.317 (0.306) data 0.001 (0.007) eta 0:47:28 loss 1.1198 (1.1036) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.222, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1525 (1.0911) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.306) data 0.001 (0.007) eta 0:47:28 loss 1.1748 (1.1091) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 838.954, TIME@all 0.305 +epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.014) eta 0:47:37 loss 1.1291 (1.0868) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:27 loss 1.1019 (1.0987) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.118, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.306) data 0.001 (0.012) eta 0:47:15 loss 1.1491 (1.0942) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1605 (1.1041) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.736, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:47:14 loss 1.1877 (1.0938) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1661 (1.1040) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 838.816, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:47:14 loss 1.2130 (1.1018) acc 100.0000 (99.8438) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:07 loss 1.1020 (1.1036) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.917, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:47:14 loss 1.1748 (1.1103) acc 96.8750 (99.5312) lr 0.026000 +epoch: [165/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1165 (1.1071) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.748, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.306) data 0.001 (0.013) eta 0:47:15 loss 1.2215 (1.0994) acc 96.8750 (99.8438) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.1276 (1.1044) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.747, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.306) data 0.000 (0.013) eta 0:47:15 loss 1.2500 (1.0975) acc 100.0000 (99.6875) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.1472 (1.1050) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.771, TIME@all 0.305 +epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:47:14 loss 1.2347 (1.0985) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1389 (1.1017) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 838.892, TIME@all 0.305 +epoch: [165/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:47:17 loss 1.2298 (1.0932) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.0918 (1.1002) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.085, TIME@all 0.305 +epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:46:48 loss 1.1262 (1.0909) acc 100.0000 (99.6875) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:47 loss 1.0959 (1.0985) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.350, TIME@all 0.305 +epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:46:47 loss 1.1228 (1.1005) acc 100.0000 (99.8438) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.1519 (1.1087) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.451, TIME@all 0.305 +epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1104 (1.0949) acc 100.0000 (99.8438) lr 0.026000 +epoch: [166/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.1347 (1.1087) acc 96.8750 (99.4531) lr 0.026000 +FPS@all 840.538, TIME@all 0.305 +epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1610 (1.0939) acc 96.8750 (99.8438) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.0716 (1.1095) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 840.387, TIME@all 0.305 +epoch: [166/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:46:47 loss 1.1080 (1.0973) acc 100.0000 (99.6875) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.1966 (1.1024) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 840.436, TIME@all 0.305 +epoch: [166/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:46:48 loss 1.1534 (1.0921) acc 96.8750 (99.8438) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.0937 (1.1037) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.389, TIME@all 0.305 +epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1113 (1.0889) acc 100.0000 (100.0000) lr 0.026000 +epoch: [166/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.0982 (1.1063) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.580, TIME@all 0.305 +epoch: [166/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:46:49 loss 1.1144 (1.0958) acc 100.0000 (100.0000) lr 0.026000 +epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:46:47 loss 1.1254 (1.1113) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.605, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:42 loss 1.0719 (1.0874) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.0977 (1.0986) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.965, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:46:41 loss 1.0804 (1.0844) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:46:38 loss 1.0963 (1.0968) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.069, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:46:41 loss 1.1135 (1.0868) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:46:39 loss 1.1296 (1.0972) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 838.950, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.1075 (1.0908) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.1278 (1.1017) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.976, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0941 (1.0884) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.0885 (1.0997) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.994, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0891 (1.0861) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0963 (1.0971) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.311, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0792 (1.0870) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0841 (1.0978) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.116, TIME@all 0.305 +epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0773 (1.0833) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0666 (1.0948) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.170, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.1378 (1.0963) acc 100.0000 (99.3750) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0695 (1.0993) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 839.554, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.1198 (1.0930) acc 100.0000 (100.0000) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0955 (1.0968) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.512, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.0802 (1.0915) acc 100.0000 (99.6875) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.1026 (1.1002) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.582, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1020 (1.0942) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0829 (1.1059) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 839.701, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1132 (1.0854) acc 100.0000 (100.0000) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0747 (1.0946) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.552, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1289 (1.0933) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.2008 (1.1116) acc 96.8750 (99.4531) lr 0.026000 +FPS@all 839.718, TIME@all 0.305 +epoch: [168/350][20/50] time 0.303 (0.305) data 0.001 (0.013) eta 0:46:22 loss 1.1536 (1.0944) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.1337 (1.1056) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 840.005, TIME@all 0.305 +epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.0894 (1.0886) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0891 (1.0973) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.560, TIME@all 0.305 +epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.013) eta 0:46:27 loss 1.0998 (1.0928) acc 100.0000 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0767 (1.1001) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.146, TIME@all 0.306 +epoch: [169/350][20/50] time 0.307 (0.307) data 0.000 (0.012) eta 0:46:27 loss 1.1573 (1.0861) acc 96.8750 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:46:16 loss 1.1870 (1.0983) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 837.078, TIME@all 0.306 +epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.1018 (1.0839) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0929 (1.0941) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.127, TIME@all 0.306 +epoch: [169/350][20/50] time 0.307 (0.307) data 0.000 (0.013) eta 0:46:27 loss 1.0693 (1.0858) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.298 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0952 (1.0964) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.060, TIME@all 0.306 +epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.0695 (1.0761) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.1129 (1.0953) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.118, TIME@all 0.306 +epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:26 loss 1.0952 (1.0832) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:15 loss 1.0939 (1.0918) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.270, TIME@all 0.306 +epoch: [169/350][20/50] time 0.305 (0.307) data 0.001 (0.014) eta 0:46:28 loss 1.0936 (1.0807) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.303 (0.307) data 0.001 (0.007) eta 0:46:17 loss 1.1271 (1.0974) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 837.257, TIME@all 0.306 +epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.0717 (1.0797) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:15 loss 1.1639 (1.0968) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.239, TIME@all 0.306 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.012) eta 0:46:12 loss 1.1162 (1.0818) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:45:57 loss 1.0990 (1.0992) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 837.766, TIME@all 0.306 +epoch: [170/350][20/50] time 0.318 (0.307) data 0.000 (0.013) eta 0:46:13 loss 1.1452 (1.0869) acc 96.8750 (99.6875) lr 0.026000 +epoch: [170/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1139 (1.0972) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.802, TIME@all 0.306 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.012) eta 0:46:11 loss 1.1059 (1.0835) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:45:57 loss 1.1138 (1.0954) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.868, TIME@all 0.306 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:11 loss 1.1278 (1.0845) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.001 (0.007) eta 0:45:56 loss 1.0907 (1.0969) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 837.971, TIME@all 0.305 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:12 loss 1.1013 (1.0827) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.0953 (1.1009) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.801, TIME@all 0.306 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:12 loss 1.0962 (1.0833) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.0701 (1.0950) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.801, TIME@all 0.306 +epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:11 loss 1.1262 (1.0788) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1872 (1.0989) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 837.941, TIME@all 0.306 +epoch: [170/350][20/50] time 0.313 (0.307) data 0.000 (0.013) eta 0:46:10 loss 1.1227 (1.0890) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1616 (1.0984) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 838.142, TIME@all 0.305 +epoch: [171/350][20/50] time 0.302 (0.305) data 0.001 (0.012) eta 0:45:35 loss 1.1189 (1.0858) acc 100.0000 (100.0000) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.1045 (1.1018) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.831, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:36 loss 1.1699 (1.0942) acc 96.8750 (99.6875) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0770 (1.1079) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 839.786, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.0826 (1.0909) acc 100.0000 (99.6875) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.1174 (1.1070) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.753, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.0856 (1.0823) acc 100.0000 (100.0000) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0766 (1.1044) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.787, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:35 loss 1.1195 (1.0864) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.0900 (1.1030) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.950, TIME@all 0.305 +epoch: [171/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.1203 (1.0843) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0788 (1.0991) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.066, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:35 loss 1.1141 (1.0858) acc 100.0000 (99.6875) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.1147 (1.1006) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.912, TIME@all 0.305 +epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:36 loss 1.1432 (1.0887) acc 100.0000 (100.0000) lr 0.026000 +epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.0842 (1.1099) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.774, TIME@all 0.305 +epoch: [172/350][20/50] time 0.293 (0.304) data 0.000 (0.012) eta 0:45:14 loss 1.1218 (1.0887) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:45:20 loss 1.1246 (1.0988) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.852, TIME@all 0.305 +epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1020 (1.0889) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:21 loss 1.1253 (1.1008) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.781, TIME@all 0.305 +epoch: [172/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:45:17 loss 1.1049 (1.0969) acc 100.0000 (99.6875) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.001 (0.007) eta 0:45:20 loss 1.0831 (1.0963) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.769, TIME@all 0.305 +epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1310 (1.0907) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:21 loss 1.1488 (1.0994) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.753, TIME@all 0.305 +epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.0892 (1.0879) acc 100.0000 (99.8438) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:20 loss 1.1144 (1.0949) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.956, TIME@all 0.305 +epoch: [172/350][20/50] time 0.293 (0.304) data 0.000 (0.012) eta 0:45:14 loss 1.1775 (1.0923) acc 96.8750 (99.6875) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:45:21 loss 1.1429 (1.0999) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 839.779, TIME@all 0.305 +epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.0977 (1.0822) acc 100.0000 (99.8438) lr 0.026000 +epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:20 loss 1.1172 (1.0915) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.901, TIME@all 0.305 +epoch: [172/350][20/50] time 0.295 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1533 (1.0874) acc 96.8750 (99.6875) lr 0.026000 +epoch: [172/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:20 loss 1.1105 (1.1049) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.058, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1816 (1.0811) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1557 (1.0985) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.688, TIME@all 0.305 +epoch: [173/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:45:05 loss 1.2135 (1.0828) acc 96.8750 (99.8438) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1162 (1.0934) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.707, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.0829 (1.0802) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.0891 (1.0958) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.763, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.0905 (1.0780) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:01 loss 1.1273 (1.0958) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.912, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.1001 (1.0830) acc 100.0000 (99.8438) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1184 (1.0976) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.676, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.1035 (1.0846) acc 100.0000 (99.8438) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1540 (1.0963) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.840, TIME@all 0.305 +epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1219 (1.0823) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1681 (1.1005) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.687, TIME@all 0.305 +epoch: [173/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1594 (1.0793) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1221 (1.0975) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.098, TIME@all 0.305 +epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:44:52 loss 1.0816 (1.0831) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:44:52 loss 1.1398 (1.0978) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.094, TIME@all 0.305 +epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0836 (1.0806) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:51 loss 1.0993 (1.0946) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.145, TIME@all 0.305 +epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:44:52 loss 1.0747 (1.0822) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.306 (0.306) data 0.001 (0.006) eta 0:44:51 loss 1.1181 (1.0923) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.158, TIME@all 0.305 +epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.1088 (1.0859) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:44:52 loss 1.0770 (1.0983) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.108, TIME@all 0.305 +epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0823 (1.0867) acc 100.0000 (99.5312) lr 0.026000 +epoch: [174/350][40/50] time 0.307 (0.306) data 0.001 (0.007) eta 0:44:51 loss 1.1231 (1.0967) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 839.123, TIME@all 0.305 +epoch: [174/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:44:48 loss 1.0921 (1.0811) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:44:50 loss 1.0759 (1.0923) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.768, TIME@all 0.305 +epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0777 (1.0831) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:44:51 loss 1.1997 (1.1017) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 839.311, TIME@all 0.305 +epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0720 (1.0876) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.307 (0.306) data 0.001 (0.007) eta 0:44:51 loss 1.1266 (1.0976) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.244, TIME@all 0.305 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:51 loss 1.0783 (1.0819) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0915 (1.0855) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.209, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0831 (1.0754) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0849 (1.0868) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 836.994, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0644 (1.0833) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0696 (1.0941) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.004, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.012) eta 0:44:52 loss 1.0838 (1.0808) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.296 (0.306) data 0.000 (0.006) eta 0:44:44 loss 1.0690 (1.0874) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.057, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0903 (1.0800) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0822 (1.0860) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.026, TIME@all 0.306 +epoch: [175/350][20/50] time 0.306 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0688 (1.0774) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.300 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0710 (1.0929) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.329, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.014) eta 0:44:52 loss 1.0835 (1.0798) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0606 (1.0902) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.148, TIME@all 0.306 +epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:52 loss 1.0902 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0882 (1.0891) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.204, TIME@all 0.306 +epoch: [176/350][20/50] time 0.315 (0.307) data 0.001 (0.013) eta 0:44:39 loss 1.1476 (1.0873) acc 96.8750 (99.8438) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0879 (1.0960) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.965, TIME@all 0.306 +epoch: [176/350][20/50] time 0.314 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0985 (1.0879) acc 100.0000 (99.5312) lr 0.026000 +epoch: [176/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0990 (1.0960) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.015, TIME@all 0.305 +epoch: [176/350][20/50] time 0.314 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0953 (1.0797) acc 100.0000 (100.0000) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1000 (1.0968) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.972, TIME@all 0.305 +epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0923 (1.0852) acc 100.0000 (99.6875) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1039 (1.0962) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 838.042, TIME@all 0.305 +epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0844 (1.0767) acc 100.0000 (100.0000) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.2082 (1.0915) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 837.971, TIME@all 0.305 +epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.014) eta 0:44:39 loss 1.1010 (1.0848) acc 100.0000 (99.8438) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0924 (1.0995) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 838.162, TIME@all 0.305 +epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.014) eta 0:44:39 loss 1.1204 (1.0879) acc 100.0000 (99.6875) lr 0.026000 +epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1311 (1.0962) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.104, TIME@all 0.305 +epoch: [176/350][20/50] time 0.311 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0779 (1.0862) acc 100.0000 (100.0000) lr 0.026000 +epoch: [176/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:44:26 loss 1.1732 (1.0959) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 838.207, TIME@all 0.305 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:44:00 loss 1.0938 (1.0850) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:43:59 loss 1.1017 (1.0953) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.794, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:44:00 loss 1.1266 (1.0863) acc 100.0000 (99.8438) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.1223 (1.0937) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.842, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:43:59 loss 1.1261 (1.0812) acc 100.0000 (99.8438) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.0937 (1.0960) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.991, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:44:00 loss 1.2402 (1.0837) acc 93.7500 (99.5312) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.0905 (1.0990) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 840.785, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.1264 (1.0785) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:43:59 loss 1.1582 (1.0972) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.806, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.0955 (1.0961) acc 100.0000 (99.3750) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.1026 (1.0975) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.945, TIME@all 0.304 +epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.1564 (1.0841) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.1819 (1.0958) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 840.842, TIME@all 0.304 +epoch: [177/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:44:00 loss 1.1029 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.0826 (1.0867) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.198, TIME@all 0.304 +epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:43:59 loss 1.0843 (1.0897) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0918 (1.1021) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 840.205, TIME@all 0.305 +epoch: [178/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.1168 (1.0855) acc 100.0000 (99.6875) lr 0.026000 +epoch: [178/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0848 (1.0928) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.339, TIME@all 0.305 +epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.0849 (1.0871) acc 100.0000 (100.0000) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0972 (1.0947) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.255, TIME@all 0.305 +epoch: [178/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.0930 (1.0831) acc 100.0000 (100.0000) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0687 (1.0890) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.236, TIME@all 0.305 +epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.0773 (1.0884) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0994 (1.0999) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.254, TIME@all 0.305 +epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1762 (1.0902) acc 96.8750 (99.5312) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0863 (1.1008) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.416, TIME@all 0.305 +epoch: [178/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1203 (1.0852) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0876 (1.0985) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.575, TIME@all 0.305 +epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1933 (1.0874) acc 100.0000 (99.6875) lr 0.026000 +epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0880 (1.1002) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 840.373, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1234 (1.0828) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1336 (1.0887) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.362, TIME@all 0.305 +epoch: [179/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1101 (1.0748) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1895 (1.0930) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 840.314, TIME@all 0.305 +epoch: [179/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1265 (1.0793) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.0877 (1.0907) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.381, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.1044 (1.0797) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.0894 (1.0863) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.528, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0925 (1.0756) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.2154 (1.0915) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.320, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0853 (1.0881) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1472 (1.0952) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.342, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0901 (1.0787) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1053 (1.0909) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.480, TIME@all 0.305 +epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0974 (1.0749) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1647 (1.0940) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 840.684, TIME@all 0.305 +epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.011) eta 0:43:20 loss 1.1295 (1.0868) acc 100.0000 (99.8438) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.1582 (1.0974) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.437, TIME@all 0.305 +epoch: [180/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:43:14 loss 1.2004 (1.0887) acc 96.8750 (99.8438) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:14 loss 1.0862 (1.1024) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 841.163, TIME@all 0.304 +epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1916 (1.0899) acc 96.8750 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.0823 (1.0969) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.458, TIME@all 0.305 +epoch: [180/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 0:43:18 loss 1.0948 (1.0833) acc 100.0000 (99.5312) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:16 loss 1.1251 (1.0953) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 840.719, TIME@all 0.305 +epoch: [180/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1775 (1.0899) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:16 loss 1.0658 (1.0946) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.600, TIME@all 0.305 +epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1046 (1.0907) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.0815 (1.0913) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.428, TIME@all 0.305 +epoch: [180/350][20/50] time 0.300 (0.304) data 0.001 (0.012) eta 0:43:17 loss 1.0795 (1.0847) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:43:14 loss 1.1400 (1.0990) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 841.154, TIME@all 0.304 +epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:43:20 loss 1.0982 (1.0860) acc 100.0000 (99.8438) lr 0.026000 +epoch: [180/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:43:16 loss 1.0839 (1.0947) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.669, TIME@all 0.305 +epoch: [181/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:43:02 loss 1.0825 (1.0916) acc 100.0000 (99.6875) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:42:54 loss 1.0889 (1.0992) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 842.329, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:43:01 loss 1.1064 (1.0889) acc 100.0000 (100.0000) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1264 (1.0984) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.433, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:43:01 loss 1.1146 (1.0857) acc 100.0000 (100.0000) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1273 (1.0963) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.552, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:43:02 loss 1.0627 (1.0895) acc 100.0000 (99.8438) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:42:54 loss 1.0977 (1.0925) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.348, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:43:02 loss 1.0741 (1.0887) acc 100.0000 (100.0000) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:42:54 loss 1.0908 (1.0992) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.352, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:43:01 loss 1.0761 (1.0911) acc 100.0000 (99.6875) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1602 (1.0938) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.496, TIME@all 0.304 +epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:43:01 loss 1.1279 (1.0923) acc 100.0000 (99.8438) lr 0.026000 +epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1315 (1.0941) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.373, TIME@all 0.304 +epoch: [181/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:43:02 loss 1.0997 (1.0975) acc 100.0000 (99.8438) lr 0.026000 +epoch: [181/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1215 (1.0933) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.630, TIME@all 0.304 +epoch: [182/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:42:53 loss 1.1315 (1.0749) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1302 (1.0847) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.382, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:53 loss 1.0820 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1499 (1.0840) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.473, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:53 loss 1.0878 (1.0779) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:44 loss 1.1183 (1.0847) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 840.639, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:54 loss 1.2511 (1.0820) acc 96.8750 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0935 (1.0866) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.392, TIME@all 0.305 +epoch: [182/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:42:54 loss 1.0980 (1.0756) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1806 (1.0906) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 840.446, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:54 loss 1.0899 (1.0757) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:42:45 loss 1.1181 (1.0851) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.408, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:53 loss 1.1017 (1.0758) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0994 (1.0880) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.563, TIME@all 0.305 +epoch: [182/350][20/50] time 0.302 (0.305) data 0.001 (0.013) eta 0:42:53 loss 1.1058 (1.0822) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0831 (1.0855) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.829, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.0898 (1.0774) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0816 (1.0859) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 842.097, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.1141 (1.0894) acc 100.0000 (99.6875) lr 0.026000 +epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.1966 (1.0990) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 842.147, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:35 loss 1.1022 (1.0929) acc 100.0000 (99.8438) lr 0.026000 +epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.1067 (1.0951) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.282, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.0924 (1.0953) acc 100.0000 (99.5312) lr 0.026000 +epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.1118 (1.0986) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.149, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:35 loss 1.0879 (1.0873) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.0987 (1.0960) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 842.320, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:36 loss 1.1427 (1.0852) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.1185 (1.0944) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.124, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:35 loss 1.1470 (1.1060) acc 100.0000 (99.6875) lr 0.026000 +epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0990 (1.1068) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.498, TIME@all 0.304 +epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.1909 (1.0989) acc 100.0000 (99.6875) lr 0.026000 +epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0890 (1.0994) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.115, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:42:20 loss 1.0758 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.006) eta 0:42:15 loss 1.1561 (1.0894) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 841.428, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:42:20 loss 1.0826 (1.0784) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.006) eta 0:42:15 loss 1.1021 (1.0866) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.491, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.1034 (1.0818) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.1377 (1.0914) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 841.460, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0977 (1.0763) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.1115 (1.0902) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.453, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0742 (1.0788) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.0715 (1.0845) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.502, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0855 (1.0769) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.0815 (1.0865) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.745, TIME@all 0.304 +epoch: [184/350][20/50] time 0.303 (0.305) data 0.001 (0.014) eta 0:42:20 loss 1.0935 (1.0787) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.007) eta 0:42:14 loss 1.0914 (1.0867) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.572, TIME@all 0.304 +epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0689 (1.0774) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:14 loss 1.1037 (1.0888) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.592, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:42:03 loss 1.1332 (1.0764) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:41:54 loss 1.0604 (1.0881) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 842.940, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:42:03 loss 1.1129 (1.0811) acc 100.0000 (99.6875) lr 0.026000 +epoch: [185/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:41:54 loss 1.0830 (1.0952) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.882, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.0645 (1.0779) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0624 (1.0922) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.944, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.1110 (1.0806) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0781 (1.0926) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.892, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.012) eta 0:42:03 loss 1.1449 (1.0784) acc 96.8750 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:41:54 loss 1.0832 (1.0891) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.894, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.013) eta 0:42:03 loss 1.1921 (1.0819) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:41:53 loss 1.0601 (1.0963) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.112, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.0882 (1.0811) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:41:53 loss 1.0668 (1.0914) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.035, TIME@all 0.304 +epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.013) eta 0:42:03 loss 1.0927 (1.0848) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0713 (1.0951) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.217, TIME@all 0.304 +epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:41:41 loss 1.1244 (1.0743) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:41:37 loss 1.0641 (1.0859) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.076, TIME@all 0.304 +epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0935 (1.0737) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0593 (1.0848) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.173, TIME@all 0.304 +epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0785 (1.0685) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0663 (1.0835) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.189, TIME@all 0.304 +epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1074 (1.0759) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0680 (1.0844) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.109, TIME@all 0.304 +epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0837 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0792 (1.0836) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.280, TIME@all 0.304 +epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:41:40 loss 1.1047 (1.0785) acc 100.0000 (99.6875) lr 0.026000 +epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:36 loss 1.0707 (1.0880) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.334, TIME@all 0.304 +epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1067 (1.0828) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0587 (1.0880) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.139, TIME@all 0.304 +epoch: [186/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1064 (1.0738) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 0:41:36 loss 1.0659 (1.0814) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.617, TIME@all 0.303 +epoch: [187/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:41:39 loss 1.1358 (1.0806) acc 96.8750 (99.8438) lr 0.026000 +epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:41:30 loss 1.0969 (1.0876) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.261, TIME@all 0.305 +epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1148 (1.0778) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1866 (1.0903) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.313, TIME@all 0.305 +epoch: [187/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1139 (1.0818) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1103 (1.0906) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.339, TIME@all 0.305 +epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1115 (1.0802) acc 96.8750 (99.6875) lr 0.026000 +epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:30 loss 1.2085 (1.0901) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.302, TIME@all 0.305 +epoch: [187/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:41:39 loss 1.1633 (1.0908) acc 96.8750 (99.8438) lr 0.026000 +epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:41:30 loss 1.1529 (1.0916) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.277, TIME@all 0.305 +epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:41:38 loss 1.1273 (1.0851) acc 100.0000 (99.8438) lr 0.026000 +epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.2969 (1.0920) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 840.464, TIME@all 0.305 +epoch: [187/350][20/50] time 0.303 (0.305) data 0.001 (0.014) eta 0:41:38 loss 1.0919 (1.0798) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1721 (1.0895) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 840.414, TIME@all 0.305 +epoch: [187/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.0984 (1.0829) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.0884 (1.0881) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.634, TIME@all 0.305 +epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:15 loss 1.1137 (1.0771) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.0762 (1.0858) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.684, TIME@all 0.305 +epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:15 loss 1.0816 (1.0748) acc 100.0000 (99.8438) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.1035 (1.0870) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.730, TIME@all 0.304 +epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:16 loss 1.0879 (1.0708) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.0671 (1.0837) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.629, TIME@all 0.305 +epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:16 loss 1.1108 (1.0768) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.1025 (1.0843) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.649, TIME@all 0.305 +epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:16 loss 1.0714 (1.0757) acc 100.0000 (99.8438) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0663 (1.0889) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.658, TIME@all 0.305 +epoch: [188/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:15 loss 1.0766 (1.0687) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0740 (1.0817) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.855, TIME@all 0.304 +epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:15 loss 1.0751 (1.0830) acc 100.0000 (99.8438) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:41:14 loss 1.1053 (1.0964) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.980, TIME@all 0.304 +epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:15 loss 1.1182 (1.0757) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0728 (1.0902) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.802, TIME@all 0.304 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.1911 (1.0755) acc 96.8750 (99.6875) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.0663 (1.0898) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.173, TIME@all 0.304 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.0824 (1.0798) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:40:49 loss 1.1035 (1.0925) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.227, TIME@all 0.304 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.0920 (1.0772) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.0680 (1.0873) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.188, TIME@all 0.304 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.0939 (1.0768) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:50 loss 1.0706 (1.0869) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.145, TIME@all 0.304 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.1769 (1.0824) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:40:49 loss 1.0679 (1.0904) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.348, TIME@all 0.304 +epoch: [189/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:40:57 loss 1.1296 (1.0735) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.1038 (1.0838) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 843.520, TIME@all 0.303 +epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.1004 (1.0691) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:49 loss 1.0810 (1.0910) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.321, TIME@all 0.304 +epoch: [189/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:40:58 loss 1.1205 (1.0781) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:50 loss 1.0726 (1.0976) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.155, TIME@all 0.304 +epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0997 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.1232 (1.0870) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.194, TIME@all 0.304 +epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0759 (1.0741) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.1019 (1.0856) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.265, TIME@all 0.304 +epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0521 (1.0749) acc 100.0000 (99.6875) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:41 loss 1.1127 (1.0861) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.221, TIME@all 0.304 +epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0593 (1.0714) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:41 loss 1.1249 (1.0884) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.194, TIME@all 0.304 +epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0689 (1.0694) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:40 loss 1.1196 (1.0801) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 842.396, TIME@all 0.304 +epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0892 (1.0739) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:40 loss 1.0838 (1.0822) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 842.330, TIME@all 0.304 +epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0738 (1.0747) acc 100.0000 (99.8438) lr 0.026000 +epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.0921 (1.0826) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.023, TIME@all 0.304 +epoch: [190/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0892 (1.0713) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.0775 (1.0849) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.489, TIME@all 0.304 +epoch: [191/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:22 loss 1.0762 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1390 (1.0815) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.239, TIME@all 0.304 +epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.2040 (1.0854) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1337 (1.0900) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.271, TIME@all 0.304 +epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.1190 (1.0815) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1556 (1.0932) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 843.305, TIME@all 0.304 +epoch: [191/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:21 loss 1.0775 (1.0721) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1615 (1.0819) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.265, TIME@all 0.304 +epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.1098 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.0990 (1.0889) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.271, TIME@all 0.304 +epoch: [191/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:40:21 loss 1.0822 (1.0749) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:40:19 loss 1.1616 (1.0908) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.452, TIME@all 0.304 +epoch: [191/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:40:21 loss 1.0906 (1.0733) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:40:19 loss 1.1432 (1.0813) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.417, TIME@all 0.304 +epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.0961 (1.0755) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1551 (1.0904) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 843.560, TIME@all 0.303 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:13 loss 1.0720 (1.0775) acc 100.0000 (99.6875) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.006) eta 0:40:07 loss 1.0744 (1.0804) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.929, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:12 loss 1.0866 (1.0703) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 0:40:07 loss 1.0975 (1.0788) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.961, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:13 loss 1.0647 (1.0708) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.1359 (1.0900) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.938, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:12 loss 1.0649 (1.0633) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.0893 (1.0813) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.154, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:13 loss 1.1015 (1.0683) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.2211 (1.0875) acc 93.7500 (99.7656) lr 0.026000 +FPS@all 841.954, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:12 loss 1.0979 (1.0701) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.1018 (1.0821) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.367, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.001 (0.014) eta 0:40:12 loss 1.0852 (1.0719) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.0892 (1.0889) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.137, TIME@all 0.304 +epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:12 loss 1.0676 (1.0685) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:40:08 loss 1.1376 (1.0890) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.974, TIME@all 0.304 +epoch: [193/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0706 (1.0697) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1734 (1.0817) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 840.206, TIME@all 0.305 +epoch: [193/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:40:10 loss 1.0917 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1410 (1.0838) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 840.306, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:10 loss 1.0568 (1.0699) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.001 (0.007) eta 0:39:57 loss 1.0762 (1.0810) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.424, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0924 (1.0750) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.0833 (1.0757) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.219, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:11 loss 1.0766 (1.0805) acc 100.0000 (99.8438) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:39:58 loss 1.1147 (1.0863) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.243, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:10 loss 1.0828 (1.0716) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:39:58 loss 1.1360 (1.0827) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 840.342, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0765 (1.0743) acc 100.0000 (99.8438) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1021 (1.0830) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.212, TIME@all 0.305 +epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:10 loss 1.1034 (1.0792) acc 96.8750 (99.8438) lr 0.026000 +epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1485 (1.0850) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.512, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:39:54 loss 1.0764 (1.0712) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:44 loss 1.0859 (1.0786) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.086, TIME@all 0.305 +epoch: [194/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0769 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.1518 (1.0819) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.115, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:39:54 loss 1.0862 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0691 (1.0857) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.149, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.001 (0.013) eta 0:39:53 loss 1.1327 (1.0784) acc 100.0000 (99.8438) lr 0.026000 +epoch: [194/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:39:43 loss 1.0631 (1.0872) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.290, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.1074 (1.0802) acc 100.0000 (99.8438) lr 0.026000 +epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0865 (1.0831) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.085, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0850 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.1142 (1.0794) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.091, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:53 loss 1.1072 (1.0743) acc 96.8750 (99.8438) lr 0.026000 +epoch: [194/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0828 (1.0843) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.245, TIME@all 0.305 +epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0740 (1.0744) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0729 (1.0821) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.411, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:39:34 loss 1.0978 (1.0745) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:39:26 loss 1.1371 (1.0859) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 840.469, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.0768 (1.0716) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.0714 (1.0792) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.510, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1576 (1.0793) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1347 (1.0925) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.493, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1330 (1.0726) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1062 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.650, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.0784 (1.0746) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1443 (1.0883) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 840.482, TIME@all 0.305 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1254 (1.0777) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1343 (1.0875) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.692, TIME@all 0.305 +epoch: [195/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1309 (1.0802) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1336 (1.0955) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.858, TIME@all 0.304 +epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1697 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1066 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.521, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.0713 (1.0798) acc 100.0000 (99.8438) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.306) data 0.000 (0.006) eta 0:39:15 loss 1.1081 (1.0856) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.022, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.1024 (1.0722) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.1029 (1.0882) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.043, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.1102 (1.0710) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.0579 (1.0860) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.065, TIME@all 0.305 +epoch: [196/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.0712 (1.0742) acc 100.0000 (99.8438) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.1236 (1.0858) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.029, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:17 loss 1.1019 (1.0776) acc 100.0000 (99.8438) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.0874 (1.0841) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.162, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:17 loss 1.0673 (1.0733) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.1012 (1.0816) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.209, TIME@all 0.305 +epoch: [196/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:39:18 loss 1.0744 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.323 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.0970 (1.0854) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.385, TIME@all 0.305 +epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:18 loss 1.1490 (1.0835) acc 96.8750 (99.3750) lr 0.026000 +epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:15 loss 1.1000 (1.0923) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.004, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0633 (1.0776) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0923 (1.0889) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.789, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0898 (1.0820) acc 100.0000 (99.6875) lr 0.026000 +epoch: [197/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.1071 (1.0949) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 839.891, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:10 loss 1.0657 (1.0700) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0859 (1.0967) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 839.767, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0676 (1.0760) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0819 (1.0881) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.795, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:10 loss 1.0701 (1.0740) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:57 loss 1.1597 (1.0882) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 839.814, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:09 loss 1.0605 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.0773 (1.0890) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.967, TIME@all 0.305 +epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:09 loss 1.0583 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.0884 (1.0827) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.935, TIME@all 0.305 +epoch: [197/350][20/50] time 0.322 (0.306) data 0.000 (0.013) eta 0:39:11 loss 1.0565 (1.0743) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.1044 (1.0906) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.139, TIME@all 0.305 +epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:47 loss 1.0775 (1.0699) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:38:44 loss 1.0753 (1.0768) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.242, TIME@all 0.305 +epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:38:47 loss 1.0640 (1.0700) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0673 (1.0808) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.399, TIME@all 0.305 +epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0825 (1.0680) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:44 loss 1.0575 (1.0777) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.252, TIME@all 0.305 +epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0772 (1.0720) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0818 (1.0814) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.274, TIME@all 0.305 +epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:38:47 loss 1.1205 (1.0818) acc 100.0000 (99.6875) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0847 (1.0880) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.462, TIME@all 0.305 +epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0718 (1.0673) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:38:44 loss 1.0758 (1.0868) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.251, TIME@all 0.305 +epoch: [198/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.1091 (1.0720) acc 100.0000 (99.6875) lr 0.026000 +epoch: [198/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0737 (1.0781) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.583, TIME@all 0.305 +epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0772 (1.0739) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0923 (1.0811) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.335, TIME@all 0.305 +epoch: [199/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1300 (1.0689) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:26 loss 1.0975 (1.0821) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.821, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1368 (1.0733) acc 100.0000 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:26 loss 1.0894 (1.0849) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.835, TIME@all 0.305 +epoch: [199/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1427 (1.0710) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:25 loss 1.1172 (1.0884) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.926, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.1263 (1.0702) acc 100.0000 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1187 (1.0849) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.841, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.2185 (1.0727) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1371 (1.0894) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.864, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:38:27 loss 1.2002 (1.0787) acc 96.8750 (99.6875) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:38:25 loss 1.1862 (1.0901) acc 96.8750 (99.4531) lr 0.026000 +FPS@all 840.042, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:38:28 loss 1.1273 (1.0732) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:25 loss 1.1778 (1.0901) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 839.981, TIME@all 0.305 +epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.2240 (1.0768) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1324 (1.0835) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.157, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:38:19 loss 1.0734 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.1385 (1.0881) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.753, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:18 loss 1.1050 (1.0825) acc 100.0000 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.1325 (1.0899) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 838.788, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:19 loss 1.1591 (1.0749) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.0606 (1.0908) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.807, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:19 loss 1.1619 (1.0907) acc 96.8750 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:38:16 loss 1.1029 (1.0899) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.747, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:19 loss 1.1258 (1.0781) acc 100.0000 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.306 (0.306) data 0.001 (0.006) eta 0:38:16 loss 1.0704 (1.0875) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.789, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:18 loss 1.1974 (1.0820) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.306 (0.306) data 0.001 (0.007) eta 0:38:16 loss 1.0995 (1.0961) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.955, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:18 loss 1.0835 (1.0804) acc 100.0000 (99.6875) lr 0.026000 +epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:38:16 loss 1.1183 (1.0909) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.899, TIME@all 0.305 +epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:17 loss 1.0746 (1.0856) acc 100.0000 (99.6875) lr 0.026000 +epoch: [200/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:38:15 loss 1.1502 (1.0932) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 839.256, TIME@all 0.305 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1089 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0668 (1.0827) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.159, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.0999 (1.0773) acc 100.0000 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:52 loss 1.0658 (1.0896) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 841.194, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.0921 (1.0838) acc 100.0000 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0735 (1.0893) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.155, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:38:01 loss 1.0928 (1.0802) acc 100.0000 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.001 (0.007) eta 0:37:52 loss 1.0699 (1.0909) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 841.355, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1514 (1.0760) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0702 (1.0856) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.175, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1548 (1.0805) acc 96.8750 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0718 (1.0922) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.504, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:38:02 loss 1.1179 (1.0730) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:37:52 loss 1.0854 (1.0886) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.222, TIME@all 0.304 +epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:01 loss 1.0994 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:52 loss 1.0724 (1.0899) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.339, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0810 (1.0721) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0675 (1.0819) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.431, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0757 (1.0792) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0879 (1.0827) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.457, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.1610 (1.0775) acc 96.8750 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0855 (1.0839) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.421, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:37:49 loss 1.0777 (1.0735) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:37:38 loss 1.0886 (1.0799) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.627, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.001 (0.012) eta 0:37:49 loss 1.0914 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0788 (1.0790) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.428, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0826 (1.0705) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.1080 (1.0833) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.413, TIME@all 0.304 +epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:37:49 loss 1.0607 (1.0799) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:37:38 loss 1.0655 (1.0829) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.580, TIME@all 0.304 +epoch: [202/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0817 (1.0771) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0676 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.744, TIME@all 0.304 +epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.1311 (1.0797) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.1744 (1.0866) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 840.996, TIME@all 0.304 +epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.0978 (1.0752) acc 100.0000 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.0951 (1.0830) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.034, TIME@all 0.304 +epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.0837 (1.0715) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.1510 (1.0848) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 841.041, TIME@all 0.304 +epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:37:34 loss 1.0989 (1.0769) acc 100.0000 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.007) eta 0:37:24 loss 1.1125 (1.0870) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 841.004, TIME@all 0.304 +epoch: [203/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:37:33 loss 1.1456 (1.0753) acc 96.8750 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.3520 (1.0909) acc 93.7500 (99.7656) lr 0.026000 +FPS@all 841.021, TIME@all 0.304 +epoch: [203/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.1581 (1.0795) acc 96.8750 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.007) eta 0:37:23 loss 1.1120 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.193, TIME@all 0.304 +epoch: [203/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.1652 (1.0826) acc 96.8750 (99.5312) lr 0.026000 +epoch: [203/350][40/50] time 0.309 (0.305) data 0.001 (0.007) eta 0:37:23 loss 1.0745 (1.0854) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.367, TIME@all 0.304 +epoch: [203/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.0812 (1.0651) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.312 (0.305) data 0.001 (0.007) eta 0:37:23 loss 1.0895 (1.0762) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.104, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.011) eta 0:37:18 loss 1.0770 (1.0723) acc 96.8750 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0638 (1.0853) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.645, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0595 (1.0647) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0606 (1.0802) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.678, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0769 (1.0748) acc 100.0000 (99.6875) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0619 (1.0853) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.736, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0683 (1.0737) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0533 (1.0885) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.649, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0613 (1.0757) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0857 (1.0865) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.680, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:37:18 loss 1.0793 (1.0810) acc 100.0000 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.007) eta 0:37:06 loss 1.0618 (1.0916) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.873, TIME@all 0.304 +epoch: [204/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0660 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.316 (0.305) data 0.000 (0.006) eta 0:37:08 loss 1.0600 (1.0848) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.967, TIME@all 0.304 +epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0689 (1.0670) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0680 (1.0834) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.814, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.012) eta 0:37:01 loss 1.0843 (1.0681) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0768 (1.0763) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.539, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.012) eta 0:37:01 loss 1.0716 (1.0675) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0911 (1.0749) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.589, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.013) eta 0:37:01 loss 1.1152 (1.0709) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.0665 (1.0806) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.584, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.001 (0.013) eta 0:37:01 loss 1.1040 (1.0668) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.0618 (1.0863) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 842.770, TIME@all 0.304 +epoch: [205/350][20/50] time 0.313 (0.305) data 0.001 (0.012) eta 0:37:01 loss 1.0708 (1.0630) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0869 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.594, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.001 (0.012) eta 0:37:01 loss 1.0844 (1.0669) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0615 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 842.603, TIME@all 0.304 +epoch: [205/350][20/50] time 0.316 (0.305) data 0.001 (0.012) eta 0:37:02 loss 1.0904 (1.0762) acc 100.0000 (99.8438) lr 0.026000 +epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0574 (1.0794) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.895, TIME@all 0.304 +epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.013) eta 0:37:01 loss 1.0719 (1.0704) acc 100.0000 (99.8438) lr 0.026000 +epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.1234 (1.0830) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 842.734, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0779 (1.0742) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.0735 (1.0815) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.425, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:36:35 loss 1.0721 (1.0738) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.1004 (1.0838) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.345, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0720 (1.0703) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0848 (1.0765) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.363, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0849 (1.0699) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0984 (1.0791) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.361, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:36:34 loss 1.0547 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:35 loss 1.0817 (1.0819) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.548, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0646 (1.0640) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.0709 (1.0737) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 841.362, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0917 (1.0807) acc 100.0000 (99.6875) lr 0.026000 +epoch: [206/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0642 (1.0870) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 841.716, TIME@all 0.304 +epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:36:34 loss 1.0692 (1.0694) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:35 loss 1.1298 (1.0776) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.504, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:26 loss 1.1383 (1.0760) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.0785 (1.0965) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 841.572, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.1532 (1.0702) acc 96.8750 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:22 loss 1.0614 (1.0930) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 841.585, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.0747 (1.0687) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:22 loss 1.1170 (1.0811) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.561, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:25 loss 1.1071 (1.0645) acc 96.8750 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.1403 (1.0804) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.609, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:26 loss 1.1218 (1.0722) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:36:22 loss 1.0831 (1.0869) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 841.602, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.1864 (1.0774) acc 96.8750 (99.6875) lr 0.026000 +epoch: [207/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:36:21 loss 1.1160 (1.0937) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 841.783, TIME@all 0.304 +epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.0845 (1.0708) acc 100.0000 (100.0000) lr 0.026000 +epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:21 loss 1.0914 (1.0841) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 841.728, TIME@all 0.304 +epoch: [207/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:36:26 loss 1.1531 (1.0726) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.0911 (1.0793) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 841.877, TIME@all 0.304 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 0:36:07 loss 1.0852 (1.0809) acc 100.0000 (99.8438) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:36:00 loss 1.0885 (1.0904) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.338, TIME@all 0.304 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:06 loss 1.0797 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0704 (1.0846) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.472, TIME@all 0.304 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0919 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:01 loss 1.0674 (1.0820) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 843.321, TIME@all 0.304 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0528 (1.0764) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0624 (1.0827) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 843.554, TIME@all 0.303 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:06 loss 1.0692 (1.0705) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:36:00 loss 1.1003 (1.0776) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 843.558, TIME@all 0.303 +epoch: [208/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0681 (1.0750) acc 100.0000 (99.8438) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0650 (1.0836) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.367, TIME@all 0.304 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0576 (1.0702) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0680 (1.0846) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.643, TIME@all 0.303 +epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0774 (1.0830) acc 100.0000 (99.6875) lr 0.026000 +epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0705 (1.0834) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.501, TIME@all 0.303 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.0956 (1.0710) acc 100.0000 (99.6875) lr 0.026000 +epoch: [209/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.1262 (1.0838) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 843.046, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.1344 (1.0715) acc 100.0000 (99.8438) lr 0.026000 +epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.0757 (1.0858) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 843.007, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.0894 (1.0663) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:48 loss 1.0998 (1.0803) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 842.952, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:51 loss 1.0638 (1.0679) acc 100.0000 (99.8438) lr 0.026000 +epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:48 loss 1.0755 (1.0773) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.959, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:35:51 loss 1.1042 (1.0776) acc 100.0000 (99.6875) lr 0.026000 +epoch: [209/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:47 loss 1.0753 (1.0924) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 843.159, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:51 loss 1.0639 (1.0681) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:47 loss 1.0722 (1.0878) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 842.989, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.1466 (1.0741) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.0837 (1.0814) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 843.110, TIME@all 0.304 +epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:50 loss 1.1132 (1.0712) acc 100.0000 (99.6875) lr 0.026000 +epoch: [209/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:35:47 loss 1.0590 (1.0846) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 843.382, TIME@all 0.304 +epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:39 loss 1.0783 (1.0746) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0873 (1.0837) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.391, TIME@all 0.304 +epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0534 (1.0677) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0882 (1.0772) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.428, TIME@all 0.304 +epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0638 (1.0709) acc 100.0000 (99.8438) lr 0.026000 +epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0805 (1.0791) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.427, TIME@all 0.304 +epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0520 (1.0668) acc 100.0000 (99.6875) lr 0.026000 +epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0593 (1.0751) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.426, TIME@all 0.304 +epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.1053 (1.0734) acc 100.0000 (99.8438) lr 0.026000 +epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0858 (1.0854) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 842.647, TIME@all 0.304 +epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0789 (1.0687) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0701 (1.0798) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.436, TIME@all 0.304 +epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0534 (1.0647) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0868 (1.0777) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.581, TIME@all 0.304 +epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:38 loss 1.0650 (1.0664) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:35:33 loss 1.1312 (1.0791) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 842.822, TIME@all 0.304 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:35:20 loss 1.0823 (1.0690) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0730 (1.0839) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.911, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:35:21 loss 1.0721 (1.0703) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0679 (1.0824) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.860, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:35:21 loss 1.0702 (1.0699) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0627 (1.0802) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.886, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:35:20 loss 1.0952 (1.0726) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:35:23 loss 1.0617 (1.0800) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.103, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:20 loss 1.0783 (1.0708) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0890 (1.0842) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.896, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:20 loss 1.0583 (1.0716) acc 100.0000 (99.8438) lr 0.026000 +epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0533 (1.0804) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 840.035, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:35:21 loss 1.0793 (1.0705) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0572 (1.0754) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.861, TIME@all 0.305 +epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:21 loss 1.0672 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0575 (1.0827) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.160, TIME@all 0.305 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:23 loss 1.0662 (1.0674) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1529 (1.0917) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 836.194, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:23 loss 1.0753 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1046 (1.0876) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 836.224, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:22 loss 1.0621 (1.0694) acc 100.0000 (99.6875) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1263 (1.0849) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 836.267, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0755 (1.0736) acc 100.0000 (99.8438) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.2314 (1.0939) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 836.186, TIME@all 0.306 +epoch: [212/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0907 (1.0772) acc 100.0000 (99.6875) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1121 (1.0973) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 836.211, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:22 loss 1.1018 (1.0694) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.1381 (1.0939) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 836.384, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0876 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.0771 (1.0894) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 836.341, TIME@all 0.306 +epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0554 (1.0646) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.1238 (1.0776) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 836.530, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.0775 (1.0845) acc 100.0000 (99.6875) lr 0.026000 +epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0737 (1.0908) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.563, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.0834 (1.0644) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0641 (1.0766) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.483, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.1027 (1.0762) acc 100.0000 (99.6875) lr 0.026000 +epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0733 (1.0816) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 837.549, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0908 (1.0768) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.007) eta 0:35:04 loss 1.0677 (1.0817) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.531, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0961 (1.0704) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.326 (0.307) data 0.001 (0.007) eta 0:35:04 loss 1.1024 (1.0781) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.550, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0830 (1.0681) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.326 (0.307) data 0.000 (0.007) eta 0:35:03 loss 1.0980 (1.0827) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.715, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.1029 (1.0784) acc 100.0000 (99.8438) lr 0.026000 +epoch: [213/350][40/50] time 0.326 (0.307) data 0.001 (0.007) eta 0:35:03 loss 1.1014 (1.0860) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.670, TIME@all 0.306 +epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:11 loss 1.0579 (1.0835) acc 100.0000 (99.5312) lr 0.026000 +epoch: [213/350][40/50] time 0.329 (0.307) data 0.000 (0.007) eta 0:35:03 loss 1.0611 (1.0854) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.941, TIME@all 0.306 +epoch: [214/350][20/50] time 0.324 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.2013 (1.0775) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0886 (1.0915) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.742, TIME@all 0.305 +epoch: [214/350][20/50] time 0.322 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.1080 (1.0786) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0918 (1.0845) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.756, TIME@all 0.305 +epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.011) eta 0:34:47 loss 1.1230 (1.0714) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1352 (1.0889) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.663, TIME@all 0.305 +epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.0961 (1.0806) acc 100.0000 (99.5312) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0745 (1.0928) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 838.725, TIME@all 0.305 +epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.0786 (1.0743) acc 100.0000 (99.6875) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1245 (1.0932) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 838.713, TIME@all 0.305 +epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.013) eta 0:34:46 loss 1.1067 (1.0745) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:34:38 loss 1.1209 (1.0901) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.901, TIME@all 0.305 +epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.013) eta 0:34:47 loss 1.1219 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:34:38 loss 1.1124 (1.1006) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.853, TIME@all 0.305 +epoch: [214/350][20/50] time 0.325 (0.306) data 0.000 (0.012) eta 0:34:48 loss 1.0665 (1.0772) acc 100.0000 (99.8438) lr 0.026000 +epoch: [214/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1498 (1.0960) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 838.999, TIME@all 0.305 +epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:40 loss 1.1533 (1.0800) acc 100.0000 (99.8438) lr 0.026000 +epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0859 (1.0843) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 834.900, TIME@all 0.307 +epoch: [215/350][20/50] time 0.312 (0.307) data 0.000 (0.011) eta 0:34:41 loss 1.0969 (1.0779) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0723 (1.0853) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 834.816, TIME@all 0.307 +epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.1385 (1.0822) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0696 (1.0862) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 834.884, TIME@all 0.307 +epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.1588 (1.0862) acc 100.0000 (99.8438) lr 0.026000 +epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0734 (1.0869) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 834.838, TIME@all 0.307 +epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.0705 (1.0776) acc 100.0000 (99.6875) lr 0.026000 +epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0732 (1.0860) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 834.823, TIME@all 0.307 +epoch: [215/350][20/50] time 0.312 (0.307) data 0.000 (0.013) eta 0:34:40 loss 1.0649 (1.0795) acc 100.0000 (99.8438) lr 0.026000 +epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.007) eta 0:34:36 loss 1.1552 (1.0866) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 834.978, TIME@all 0.307 +epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.013) eta 0:34:41 loss 1.1062 (1.0731) acc 100.0000 (99.8438) lr 0.026000 +epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.007) eta 0:34:36 loss 1.0801 (1.0877) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 834.948, TIME@all 0.307 +epoch: [215/350][20/50] time 0.310 (0.307) data 0.000 (0.012) eta 0:34:40 loss 1.0651 (1.0731) acc 100.0000 (99.6875) lr 0.026000 +epoch: [215/350][40/50] time 0.305 (0.307) data 0.001 (0.006) eta 0:34:36 loss 1.0648 (1.0854) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 835.187, TIME@all 0.307 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:34:13 loss 1.1136 (1.0739) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1388 (1.0858) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.563, TIME@all 0.305 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0647 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:34:07 loss 1.0818 (1.0840) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.562, TIME@all 0.305 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1195 (1.0707) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1088 (1.0862) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 840.543, TIME@all 0.305 +epoch: [216/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:34:13 loss 1.1571 (1.0782) acc 100.0000 (99.8438) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.0761 (1.0955) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 840.550, TIME@all 0.305 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1363 (1.0715) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1599 (1.0844) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 840.574, TIME@all 0.305 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0863 (1.0630) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:34:06 loss 1.0819 (1.0869) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 840.757, TIME@all 0.304 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0975 (1.0687) acc 100.0000 (99.8438) lr 0.026000 +epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:34:07 loss 1.1633 (1.0872) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 840.699, TIME@all 0.305 +epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1000 (1.0728) acc 100.0000 (99.8438) lr 0.026000 +epoch: [216/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:34:06 loss 1.1176 (1.0796) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 840.856, TIME@all 0.304 +epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:34:01 loss 1.1956 (1.0761) acc 96.8750 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:33:57 loss 1.1372 (1.0854) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.836, TIME@all 0.306 +epoch: [217/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:34:00 loss 1.1324 (1.0792) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1301 (1.0891) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.922, TIME@all 0.306 +epoch: [217/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.1098 (1.0721) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.2259 (1.0918) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 837.871, TIME@all 0.306 +epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.0762 (1.0755) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.0816 (1.0908) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.874, TIME@all 0.306 +epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.1653 (1.0855) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1260 (1.0931) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.884, TIME@all 0.306 +epoch: [217/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:34:00 loss 1.1270 (1.0772) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:57 loss 1.1082 (1.0894) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.057, TIME@all 0.305 +epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:00 loss 1.1038 (1.0757) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1269 (1.0803) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.039, TIME@all 0.305 +epoch: [217/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:34:00 loss 1.0955 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [217/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:57 loss 1.0982 (1.0858) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.313, TIME@all 0.305 +epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:33:44 loss 1.0625 (1.0712) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:33:45 loss 1.0750 (1.0790) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.088, TIME@all 0.305 +epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0515 (1.0677) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0694 (1.0821) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.070, TIME@all 0.305 +epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:33:44 loss 1.0642 (1.0718) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:33:45 loss 1.0774 (1.0805) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.162, TIME@all 0.305 +epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0675 (1.0685) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0506 (1.0752) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.094, TIME@all 0.305 +epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0937 (1.0715) acc 96.8750 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0729 (1.0771) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.096, TIME@all 0.305 +epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0630 (1.0704) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:44 loss 1.1334 (1.0859) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.277, TIME@all 0.305 +epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0969 (1.0693) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:44 loss 1.0562 (1.0813) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.214, TIME@all 0.305 +epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:33:43 loss 1.0796 (1.0717) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:33:44 loss 1.0829 (1.0868) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 838.494, TIME@all 0.305 +epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:33:41 loss 1.0757 (1.0658) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:28 loss 1.0570 (1.0759) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.622, TIME@all 0.306 +epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0672 (1.0745) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:28 loss 1.0857 (1.0869) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.649, TIME@all 0.306 +epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:33:41 loss 1.0549 (1.0699) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:27 loss 1.0654 (1.0803) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.690, TIME@all 0.306 +epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0769 (1.0727) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:28 loss 1.0623 (1.0850) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.636, TIME@all 0.306 +epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:40 loss 1.0606 (1.0645) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:27 loss 1.0706 (1.0827) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.827, TIME@all 0.306 +epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0695 (1.0670) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:33:27 loss 1.0768 (1.0848) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.769, TIME@all 0.306 +epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0857 (1.0665) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:28 loss 1.0589 (1.0837) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.629, TIME@all 0.306 +epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0602 (1.0683) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:33:27 loss 1.0589 (1.0874) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.987, TIME@all 0.305 +epoch: [220/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:33:24 loss 1.0758 (1.0767) acc 100.0000 (99.6875) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.006) eta 0:33:15 loss 1.0909 (1.0845) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.885, TIME@all 0.306 +epoch: [220/350][20/50] time 0.309 (0.307) data 0.001 (0.012) eta 0:33:24 loss 1.0693 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.006) eta 0:33:15 loss 1.0731 (1.0821) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.970, TIME@all 0.306 +epoch: [220/350][20/50] time 0.306 (0.307) data 0.001 (0.013) eta 0:33:24 loss 1.0873 (1.0737) acc 96.8750 (99.8438) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.0889 (1.0848) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.880, TIME@all 0.306 +epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0792 (1.0692) acc 100.0000 (99.8438) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.1520 (1.0837) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 837.872, TIME@all 0.306 +epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:23 loss 1.0592 (1.0676) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:14 loss 1.0703 (1.0788) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.110, TIME@all 0.305 +epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0675 (1.0653) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:15 loss 1.0984 (1.0785) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.914, TIME@all 0.306 +epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0876 (1.0683) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:14 loss 1.0830 (1.0770) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.059, TIME@all 0.305 +epoch: [220/350][20/50] time 0.308 (0.307) data 0.000 (0.013) eta 0:33:23 loss 1.1131 (1.0733) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.316 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.0848 (1.0822) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 838.322, TIME@all 0.305 +epoch: [221/350][20/50] time 0.309 (0.306) data 0.000 (0.011) eta 0:33:00 loss 1.0975 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0945 (1.0813) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.319, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1292 (1.0724) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.1764 (1.0822) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 837.375, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1014 (1.0801) acc 100.0000 (99.8438) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0786 (1.0777) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.354, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1455 (1.0768) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.1198 (1.0803) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.408, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.001 (0.013) eta 0:32:59 loss 1.1493 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:56 loss 1.1349 (1.0793) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.539, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.0901 (1.0794) acc 100.0000 (99.8438) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0819 (1.0790) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 837.368, TIME@all 0.306 +epoch: [221/350][20/50] time 0.316 (0.306) data 0.000 (0.012) eta 0:33:01 loss 1.0670 (1.0719) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.305 (0.306) data 0.001 (0.006) eta 0:32:57 loss 1.0988 (1.0774) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.672, TIME@all 0.306 +epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:32:59 loss 1.0991 (1.0730) acc 100.0000 (99.8438) lr 0.026000 +epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0865 (1.0740) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 837.500, TIME@all 0.306 +epoch: [222/350][20/50] time 0.301 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0676 (1.0747) acc 100.0000 (99.8438) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0529 (1.0845) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 838.052, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0531 (1.0679) acc 100.0000 (99.8438) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0866 (1.0780) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 838.096, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0899 (1.0648) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0828 (1.0783) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 837.999, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0808 (1.0694) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.1238 (1.0806) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 838.023, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0731 (1.0650) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0555 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 838.040, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:32:44 loss 1.0548 (1.0638) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0711 (1.0715) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 838.225, TIME@all 0.305 +epoch: [222/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0827 (1.0741) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0632 (1.0776) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 838.351, TIME@all 0.305 +epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0567 (1.0658) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.1484 (1.0834) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 838.169, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.1059 (1.0746) acc 100.0000 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.308 (0.306) data 0.001 (0.006) eta 0:32:23 loss 1.1427 (1.0850) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.078, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.1014 (1.0643) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1549 (1.0771) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.116, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.0675 (1.0661) acc 100.0000 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.309 (0.306) data 0.001 (0.006) eta 0:32:23 loss 1.1534 (1.0822) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.130, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0742 (1.0679) acc 100.0000 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:32:23 loss 1.0902 (1.0750) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.297, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.0806 (1.0624) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.308 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1209 (1.0789) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.098, TIME@all 0.305 +epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0577 (1.0621) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:32:23 loss 1.0846 (1.0768) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.248, TIME@all 0.305 +epoch: [223/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0871 (1.0658) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.308 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1732 (1.0789) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.093, TIME@all 0.305 +epoch: [223/350][20/50] time 0.298 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0736 (1.0615) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.305 (0.306) data 0.001 (0.007) eta 0:32:22 loss 1.0982 (1.0751) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 839.516, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.012) eta 0:32:15 loss 1.0589 (1.0715) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:08 loss 1.0985 (1.0884) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.133, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0697 (1.0706) acc 100.0000 (99.8438) lr 0.026000 +epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.1222 (1.0823) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.169, TIME@all 0.305 +epoch: [224/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:32:10 loss 1.0873 (1.0775) acc 100.0000 (99.8438) lr 0.026000 +epoch: [224/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:32:06 loss 1.1253 (1.0812) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 839.859, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.014) eta 0:32:15 loss 1.0685 (1.0773) acc 100.0000 (99.6875) lr 0.026000 +epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.1036 (1.0842) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 839.270, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.014) eta 0:32:15 loss 1.0675 (1.0646) acc 100.0000 (99.8438) lr 0.026000 +epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.1082 (1.0787) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 839.331, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0787 (1.0655) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.0879 (1.0822) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 839.156, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0668 (1.0692) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.1010 (1.0797) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.148, TIME@all 0.305 +epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:14 loss 1.1143 (1.0768) acc 96.8750 (99.5312) lr 0.026000 +epoch: [224/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.0911 (1.0808) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 839.511, TIME@all 0.305 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.0595 (1.0711) acc 100.0000 (99.8438) lr 0.026000 +epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.1035 (1.0860) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 836.801, TIME@all 0.306 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.1136 (1.0647) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.0814 (1.0719) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 836.828, TIME@all 0.306 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0696 (1.0622) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:31:58 loss 1.0720 (1.0801) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 836.855, TIME@all 0.306 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.1251 (1.0682) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:31:57 loss 1.0897 (1.0791) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 836.970, TIME@all 0.306 +epoch: [225/350][20/50] time 0.300 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0630 (1.0633) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:31:58 loss 1.0632 (1.0786) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 837.157, TIME@all 0.306 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.1726 (1.0669) acc 96.8750 (99.8438) lr 0.026000 +epoch: [225/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:31:57 loss 1.1251 (1.0791) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 837.007, TIME@all 0.306 +epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0557 (1.0659) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:31:58 loss 1.0921 (1.0866) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 836.835, TIME@all 0.306 +epoch: [225/350][20/50] time 0.300 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.0599 (1.0651) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.0951 (1.0837) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 836.894, TIME@all 0.306 +epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:31:50 loss 1.1594 (1.0733) acc 96.8750 (99.8438) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.006) eta 0:31:40 loss 1.0714 (1.0754) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.529, TIME@all 0.305 +epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0868 (1.0671) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.320 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0514 (1.0774) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.606, TIME@all 0.305 +epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0966 (1.0651) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0761 (1.0763) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.549, TIME@all 0.305 +epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0957 (1.0690) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0916 (1.0783) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.541, TIME@all 0.305 +epoch: [226/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0911 (1.0715) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:39 loss 1.1611 (1.0787) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.703, TIME@all 0.305 +epoch: [226/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:31:49 loss 1.1367 (1.0691) acc 96.8750 (99.8438) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:31:39 loss 1.0959 (1.0807) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 838.939, TIME@all 0.305 +epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.1645 (1.0660) acc 96.8750 (99.8438) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0686 (1.0764) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 838.542, TIME@all 0.305 +epoch: [226/350][20/50] time 0.304 (0.307) data 0.001 (0.014) eta 0:31:50 loss 1.1560 (1.0688) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.317 (0.306) data 0.001 (0.007) eta 0:31:39 loss 1.0925 (1.0800) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.744, TIME@all 0.305 +epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0731 (1.0640) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0734 (1.0678) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.632, TIME@all 0.305 +epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.012) eta 0:31:40 loss 1.0566 (1.0658) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:31:25 loss 1.0551 (1.0739) acc 100.0000 (99.4531) lr 0.002600 +FPS@all 838.625, TIME@all 0.305 +epoch: [227/350][20/50] time 0.334 (0.307) data 0.000 (0.012) eta 0:31:40 loss 1.0607 (1.0706) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:31:25 loss 1.0866 (1.0724) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.706, TIME@all 0.305 +epoch: [227/350][20/50] time 0.335 (0.307) data 0.000 (0.014) eta 0:31:40 loss 1.0682 (1.0681) acc 100.0000 (99.6875) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:24 loss 1.0831 (1.0750) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 838.813, TIME@all 0.305 +epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0969 (1.0656) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0582 (1.0694) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.617, TIME@all 0.305 +epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.1015 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:24 loss 1.0708 (1.0674) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.772, TIME@all 0.305 +epoch: [227/350][20/50] time 0.334 (0.308) data 0.001 (0.013) eta 0:31:40 loss 1.1135 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0583 (1.0698) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.628, TIME@all 0.305 +epoch: [227/350][20/50] time 0.334 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0772 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0637 (1.0713) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.923, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:31:11 loss 1.0697 (1.0616) acc 100.0000 (99.8438) lr 0.002600 +epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0901 (1.0709) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.216, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 0:31:10 loss 1.0547 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0652 (1.0663) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.181, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:31:10 loss 1.0739 (1.0675) acc 100.0000 (99.6875) lr 0.002600 +epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:31:08 loss 1.1467 (1.0760) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 838.114, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.014) eta 0:31:11 loss 1.0570 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0733 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.145, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:31:11 loss 1.0521 (1.0621) acc 100.0000 (99.8438) lr 0.002600 +epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0828 (1.0769) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 838.138, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.014) eta 0:31:10 loss 1.0678 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0852 (1.0703) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.322, TIME@all 0.305 +epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 0:31:10 loss 1.0841 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0606 (1.0772) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.287, TIME@all 0.305 +epoch: [228/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:31:10 loss 1.0642 (1.0607) acc 100.0000 (99.8438) lr 0.002600 +epoch: [228/350][40/50] time 0.308 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0561 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.439, TIME@all 0.305 +epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:30:57 loss 1.0811 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0495 (1.0620) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 837.116, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:30:57 loss 1.1091 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0478 (1.0718) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 837.091, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.011) eta 0:30:57 loss 1.1131 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0648 (1.0744) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 836.999, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:30:57 loss 1.1164 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.1143 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 837.053, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:30:57 loss 1.1093 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.0508 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 837.013, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:30:57 loss 1.1275 (1.0637) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.001 (0.007) eta 0:30:55 loss 1.0647 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 837.212, TIME@all 0.306 +epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:30:57 loss 1.0646 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.0517 (1.0716) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 837.168, TIME@all 0.306 +epoch: [229/350][20/50] time 0.313 (0.306) data 0.000 (0.012) eta 0:30:57 loss 1.1966 (1.0689) acc 96.8750 (99.8438) lr 0.002600 +epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0545 (1.0711) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 837.357, TIME@all 0.306 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.011) eta 0:30:35 loss 1.1502 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0655 (1.0739) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.128, TIME@all 0.305 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1663 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0784 (1.0726) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 839.174, TIME@all 0.305 +epoch: [230/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1142 (1.0708) acc 100.0000 (99.8438) lr 0.002600 +epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0920 (1.0777) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.221, TIME@all 0.305 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:30:35 loss 1.1536 (1.0663) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:36 loss 1.1164 (1.0745) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 839.156, TIME@all 0.305 +epoch: [230/350][20/50] time 0.301 (0.304) data 0.001 (0.012) eta 0:30:35 loss 1.1429 (1.0763) acc 100.0000 (99.6875) lr 0.002600 +epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0840 (1.0818) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 839.157, TIME@all 0.305 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:30:35 loss 1.1021 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 0:30:35 loss 1.0822 (1.0755) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.349, TIME@all 0.305 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1059 (1.0693) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0657 (1.0775) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 839.267, TIME@all 0.305 +epoch: [230/350][20/50] time 0.300 (0.304) data 0.001 (0.012) eta 0:30:35 loss 1.1046 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.1221 (1.0768) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 839.299, TIME@all 0.305 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.012) eta 0:30:37 loss 1.0546 (1.0658) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.006) eta 0:30:28 loss 1.0562 (1.0730) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 836.360, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0519 (1.0599) acc 100.0000 (99.8438) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0522 (1.0768) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 836.512, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0587 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.001 (0.007) eta 0:30:28 loss 1.0522 (1.0745) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 836.367, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0541 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0802 (1.0723) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 836.435, TIME@all 0.306 +epoch: [231/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0855 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0521 (1.0699) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 836.381, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0503 (1.0617) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0675 (1.0745) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 836.370, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:36 loss 1.0557 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.1006 (1.0725) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 836.564, TIME@all 0.306 +epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:36 loss 1.0561 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.301 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0471 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 836.738, TIME@all 0.306 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:23 loss 1.0676 (1.0596) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0716 (1.0644) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.414, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:30:22 loss 1.0512 (1.0601) acc 100.0000 (99.8438) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.006) eta 0:30:11 loss 1.0624 (1.0676) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.377, TIME@all 0.305 +epoch: [232/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:30:22 loss 1.0639 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.325 (0.307) data 0.000 (0.006) eta 0:30:11 loss 1.0913 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.491, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.014) eta 0:30:22 loss 1.0853 (1.0597) acc 100.0000 (99.8438) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.306) data 0.000 (0.007) eta 0:30:11 loss 1.0514 (1.0619) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.614, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0590 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0798 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.385, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0924 (1.0612) acc 96.8750 (99.8438) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0534 (1.0631) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.430, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0801 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.322 (0.306) data 0.000 (0.007) eta 0:30:11 loss 1.0584 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 838.556, TIME@all 0.305 +epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:23 loss 1.0841 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.324 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0857 (1.0639) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 838.652, TIME@all 0.305 +epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:29:49 loss 1.0794 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0489 (1.0722) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.839, TIME@all 0.304 +epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:49 loss 1.0637 (1.0603) acc 100.0000 (99.8438) lr 0.002600 +epoch: [233/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0529 (1.0685) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.866, TIME@all 0.304 +epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:29:48 loss 1.0836 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0628 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.932, TIME@all 0.304 +epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:49 loss 1.0668 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0547 (1.0724) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.891, TIME@all 0.304 +epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:49 loss 1.0542 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:29:43 loss 1.0552 (1.0756) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.846, TIME@all 0.304 +epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:48 loss 1.0598 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:29:42 loss 1.0473 (1.0720) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.046, TIME@all 0.304 +epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:48 loss 1.0614 (1.0655) acc 100.0000 (99.6875) lr 0.002600 +epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:29:42 loss 1.0634 (1.0749) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.006, TIME@all 0.304 +epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:48 loss 1.0649 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +epoch: [233/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0696 (1.0702) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.221, TIME@all 0.304 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0660 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0754 (1.0692) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.455, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0589 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0580 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.427, TIME@all 0.303 +epoch: [234/350][20/50] time 0.301 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0538 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0557 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.466, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0602 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0539 (1.0629) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.419, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0566 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0493 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.559, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0626 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0619 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.426, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.014) eta 0:29:22 loss 1.0579 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0507 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.631, TIME@all 0.303 +epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0531 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0510 (1.0644) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.828, TIME@all 0.303 +epoch: [235/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:29:16 loss 1.0606 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.0514 (1.0611) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.555, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:29:16 loss 1.0779 (1.0595) acc 100.0000 (99.8438) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.0585 (1.0693) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.618, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.1027 (1.0549) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0760 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.560, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.1783 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.1281 (1.0711) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.586, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.1109 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:08 loss 1.0725 (1.0674) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.775, TIME@all 0.303 +epoch: [235/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.0944 (1.0545) acc 96.8750 (99.8438) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.1023 (1.0638) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.917, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.0698 (1.0519) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0710 (1.0628) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.719, TIME@all 0.303 +epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.0480 (1.0525) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0593 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.562, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:28:59 loss 1.0690 (1.0587) acc 100.0000 (99.8438) lr 0.002600 +epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:53 loss 1.0535 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.330, TIME@all 0.303 +epoch: [236/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:29:00 loss 1.0778 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0736 (1.0719) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.312, TIME@all 0.303 +epoch: [236/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:28:59 loss 1.0646 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:53 loss 1.0536 (1.0737) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.423, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0958 (1.0631) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0504 (1.0703) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.543, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0964 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0512 (1.0702) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.327, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0558 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:28:53 loss 1.0780 (1.0683) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.337, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0823 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0621 (1.0710) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.494, TIME@all 0.303 +epoch: [236/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:28:59 loss 1.0801 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0713 (1.0704) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.640, TIME@all 0.303 +epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.1111 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0725 (1.0739) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.775, TIME@all 0.304 +epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:28:53 loss 1.0745 (1.0682) acc 100.0000 (99.8438) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0619 (1.0757) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.848, TIME@all 0.304 +epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.0845 (1.0572) acc 100.0000 (99.8438) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1064 (1.0693) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.690, TIME@all 0.304 +epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 0:28:54 loss 1.0766 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1097 (1.0724) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.716, TIME@all 0.304 +epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:28:53 loss 1.1845 (1.0620) acc 96.8750 (99.8438) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1033 (1.0731) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 841.939, TIME@all 0.304 +epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.0957 (1.0653) acc 100.0000 (99.8438) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1526 (1.0720) acc 96.8750 (99.6094) lr 0.002600 +FPS@all 841.715, TIME@all 0.304 +epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 0:28:54 loss 1.1196 (1.0688) acc 100.0000 (99.6875) lr 0.002600 +epoch: [237/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0943 (1.0708) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.108, TIME@all 0.304 +epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:28:53 loss 1.0854 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0548 (1.0712) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.851, TIME@all 0.304 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0523 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0640 (1.0709) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.160, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0537 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0483 (1.0680) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.091, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:32 loss 1.0691 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0571 (1.0746) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.200, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:28:32 loss 1.0745 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:25 loss 1.0522 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.365, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.001 (0.012) eta 0:28:32 loss 1.0656 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0690 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.545, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:32 loss 1.0582 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0600 (1.0724) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.167, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0565 (1.0578) acc 100.0000 (99.8438) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0504 (1.0642) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.136, TIME@all 0.303 +epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:28:32 loss 1.0756 (1.0616) acc 100.0000 (99.8438) lr 0.002600 +epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:25 loss 1.0660 (1.0707) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.322, TIME@all 0.303 +epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:28:16 loss 1.0573 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0578 (1.0671) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.790, TIME@all 0.304 +epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0596 (1.0561) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0551 (1.0682) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.832, TIME@all 0.304 +epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0569 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0506 (1.0719) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.841, TIME@all 0.304 +epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:15 loss 1.0488 (1.0638) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:28:10 loss 1.0572 (1.0739) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.987, TIME@all 0.304 +epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0617 (1.0620) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0607 (1.0699) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.817, TIME@all 0.304 +epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:28:16 loss 1.0530 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0509 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.808, TIME@all 0.304 +epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:15 loss 1.0568 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:28:10 loss 1.0491 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.976, TIME@all 0.304 +epoch: [239/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 0:28:15 loss 1.0490 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0495 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.258, TIME@all 0.304 +epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.012) eta 0:28:07 loss 1.0766 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:27:59 loss 1.1132 (1.0669) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 842.019, TIME@all 0.304 +epoch: [240/350][20/50] time 0.321 (0.305) data 0.000 (0.012) eta 0:28:08 loss 1.0560 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:27:58 loss 1.0743 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.140, TIME@all 0.304 +epoch: [240/350][20/50] time 0.325 (0.306) data 0.000 (0.013) eta 0:28:09 loss 1.0664 (1.0658) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:27:59 loss 1.1516 (1.0700) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 842.073, TIME@all 0.304 +epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0803 (1.0684) acc 100.0000 (99.6875) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:59 loss 1.0778 (1.0678) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.030, TIME@all 0.304 +epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.014) eta 0:28:06 loss 1.1427 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0695 (1.0696) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.248, TIME@all 0.304 +epoch: [240/350][20/50] time 0.319 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0599 (1.0689) acc 100.0000 (99.6875) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0865 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.384, TIME@all 0.304 +epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:06 loss 1.1005 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0954 (1.0709) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.193, TIME@all 0.304 +epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0707 (1.0630) acc 100.0000 (99.8438) lr 0.002600 +epoch: [240/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:27:59 loss 1.2014 (1.0689) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 842.041, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0675 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.2126 (1.0694) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 842.412, TIME@all 0.304 +epoch: [241/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0580 (1.0532) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0818 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.458, TIME@all 0.304 +epoch: [241/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0745 (1.0556) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0595 (1.0664) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.495, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0721 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0691 (1.0713) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.415, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0579 (1.0546) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0615 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.641, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.1259 (1.0616) acc 96.8750 (99.8438) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0517 (1.0719) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.423, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0646 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0900 (1.0655) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.773, TIME@all 0.304 +epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0586 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0508 (1.0671) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.549, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:27:32 loss 1.0602 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:27 loss 1.1332 (1.0653) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.036, TIME@all 0.304 +epoch: [242/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:31 loss 1.0726 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.1061 (1.0677) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.087, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:27:31 loss 1.0500 (1.0604) acc 100.0000 (99.8438) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.1129 (1.0633) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.120, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:31 loss 1.0644 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.0629 (1.0639) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.043, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:32 loss 1.0574 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:27 loss 1.0683 (1.0688) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.051, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:27:31 loss 1.0881 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.0963 (1.0654) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.238, TIME@all 0.304 +epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:31 loss 1.0757 (1.0596) acc 100.0000 (99.6875) lr 0.002600 +epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.1029 (1.0706) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 842.201, TIME@all 0.304 +epoch: [242/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:27:31 loss 1.0554 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.1041 (1.0622) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.413, TIME@all 0.304 +epoch: [243/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.0680 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0529 (1.0669) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.522, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.1270 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0539 (1.0681) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.613, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0735 (1.0637) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0575 (1.0701) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.548, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0662 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0516 (1.0696) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.555, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:13 loss 1.0554 (1.0570) acc 100.0000 (99.8438) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0708 (1.0640) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.743, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.0592 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0629 (1.0698) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.562, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0572 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0570 (1.0679) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.692, TIME@all 0.304 +epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0582 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0658 (1.0700) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.909, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0542 (1.0521) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0503 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.584, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.011) eta 0:27:03 loss 1.0894 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0839 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.664, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0748 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0835 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.576, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0504 (1.0514) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0564 (1.0601) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.591, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0925 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0597 (1.0680) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.622, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:02 loss 1.0482 (1.0518) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0797 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.991, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:02 loss 1.0543 (1.0533) acc 100.0000 (99.8438) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0697 (1.0628) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.735, TIME@all 0.304 +epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:27:02 loss 1.0674 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:26:56 loss 1.0711 (1.0620) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.796, TIME@all 0.304 +epoch: [245/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0496 (1.0723) acc 100.0000 (99.6875) lr 0.002600 +epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0612 (1.0762) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.626, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0531 (1.0631) acc 100.0000 (99.8438) lr 0.002600 +epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0560 (1.0699) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.665, TIME@all 0.304 +epoch: [245/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0500 (1.0658) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0650 (1.0673) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.677, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0514 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0590 (1.0685) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.825, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0504 (1.0598) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:26:41 loss 1.0801 (1.0711) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.632, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0571 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0615 (1.0680) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.624, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0722 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0565 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.791, TIME@all 0.304 +epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0562 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0545 (1.0703) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.944, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.012) eta 0:26:30 loss 1.0540 (1.0609) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:25 loss 1.0658 (1.0690) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.181, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:26:30 loss 1.0564 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:25 loss 1.0667 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.208, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:26:30 loss 1.0477 (1.0673) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:25 loss 1.0588 (1.0710) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.209, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:26:30 loss 1.0602 (1.0600) acc 100.0000 (99.8438) lr 0.002600 +epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:24 loss 1.0520 (1.0729) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.408, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:26:30 loss 1.0604 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:25 loss 1.0909 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.214, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.012) eta 0:26:30 loss 1.0583 (1.0615) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:24 loss 1.0700 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.251, TIME@all 0.304 +epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:26:30 loss 1.0921 (1.0717) acc 100.0000 (99.5312) lr 0.002600 +epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:24 loss 1.0718 (1.0711) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.367, TIME@all 0.304 +epoch: [246/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:26:30 loss 1.1025 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:26:24 loss 1.0799 (1.0704) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.613, TIME@all 0.303 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:26:19 loss 1.0700 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.1010 (1.0678) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 840.781, TIME@all 0.304 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:26:19 loss 1.1485 (1.0696) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0757 (1.0700) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 840.811, TIME@all 0.304 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0522 (1.0563) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0581 (1.0624) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 840.953, TIME@all 0.304 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0714 (1.0549) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0693 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 840.778, TIME@all 0.304 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:26:19 loss 1.0884 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:26:13 loss 1.0634 (1.0723) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.003, TIME@all 0.304 +epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0870 (1.0631) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.1297 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 840.802, TIME@all 0.304 +epoch: [247/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.1296 (1.0650) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0552 (1.0674) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.102, TIME@all 0.304 +epoch: [247/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.2360 (1.0637) acc 96.8750 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0858 (1.0703) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 840.828, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:25:58 loss 1.0452 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:25:53 loss 1.0537 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.056, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0517 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0831 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.118, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:57 loss 1.0450 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:52 loss 1.0692 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.089, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:57 loss 1.0481 (1.0522) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:25:52 loss 1.0498 (1.0682) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.263, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:58 loss 1.0559 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0458 (1.0615) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.059, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:57 loss 1.0702 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:52 loss 1.0579 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.218, TIME@all 0.304 +epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0488 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0542 (1.0678) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.076, TIME@all 0.304 +epoch: [248/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0499 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0530 (1.0665) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.372, TIME@all 0.304 +epoch: [249/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.0646 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:39 loss 1.0790 (1.0654) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.725, TIME@all 0.303 +epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:25:46 loss 1.0560 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:38 loss 1.0510 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.775, TIME@all 0.303 +epoch: [249/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:25:46 loss 1.0851 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:39 loss 1.0547 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.691, TIME@all 0.303 +epoch: [249/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:25:46 loss 1.0856 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:39 loss 1.0603 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.723, TIME@all 0.303 +epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.1087 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:39 loss 1.0586 (1.0637) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.734, TIME@all 0.303 +epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.0564 (1.0609) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:38 loss 1.0740 (1.0718) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.918, TIME@all 0.303 +epoch: [249/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.2058 (1.0704) acc 96.8750 (99.6875) lr 0.002600 +epoch: [249/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:25:39 loss 1.0546 (1.0744) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.987, TIME@all 0.303 +epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.1066 (1.0581) acc 96.8750 (99.8438) lr 0.002600 +epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:38 loss 1.1457 (1.0695) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.846, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.011) eta 0:25:27 loss 1.0620 (1.0636) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0700 (1.0718) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.563, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0702 (1.0671) acc 100.0000 (99.6875) lr 0.002600 +epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0596 (1.0749) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 843.571, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0827 (1.0606) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:25:24 loss 1.1223 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.581, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0488 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.1120 (1.0728) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.620, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:25:27 loss 1.0987 (1.0655) acc 100.0000 (99.6875) lr 0.002600 +epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:25:23 loss 1.0833 (1.0719) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.737, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0694 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0663 (1.0693) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.572, TIME@all 0.303 +epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0676 (1.0617) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:23 loss 1.0622 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.696, TIME@all 0.303 +epoch: [250/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:25:26 loss 1.0697 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:25:23 loss 1.0694 (1.0695) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.989, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0743 (1.0581) acc 100.0000 (99.8438) lr 0.002600 +epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.0940 (1.0701) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.013, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:25:14 loss 1.1017 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:25:08 loss 1.0687 (1.0705) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.077, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1298 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.0604 (1.0731) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.976, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1059 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1003 (1.0709) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 844.002, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1578 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1072 (1.0663) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 844.017, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0864 (1.0595) acc 100.0000 (99.8438) lr 0.002600 +epoch: [251/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:25:07 loss 1.0921 (1.0702) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.197, TIME@all 0.303 +epoch: [251/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0825 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:25:08 loss 1.0544 (1.0681) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.338, TIME@all 0.303 +epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0809 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1144 (1.0704) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.131, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0785 (1.0610) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0625 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.679, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0918 (1.0633) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.300 (0.304) data 0.001 (0.006) eta 0:24:52 loss 1.0472 (1.0632) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.719, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0556 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0503 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.759, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0664 (1.0590) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0512 (1.0637) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.883, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0954 (1.0563) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0522 (1.0612) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.676, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.1267 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0423 (1.0669) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.830, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0785 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:24:52 loss 1.0486 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.034, TIME@all 0.303 +epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0514 (1.0578) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0452 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.655, TIME@all 0.303 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:48 loss 1.1000 (1.0593) acc 100.0000 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0431 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.196, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:48 loss 1.1287 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0627 (1.0648) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.250, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.1076 (1.0586) acc 100.0000 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0566 (1.0646) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.360, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.0866 (1.0592) acc 100.0000 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0855 (1.0630) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.292, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:49 loss 1.1651 (1.0586) acc 96.8750 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0469 (1.0616) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.221, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:49 loss 1.0895 (1.0546) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0749 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.216, TIME@all 0.304 +epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.0983 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0994 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.417, TIME@all 0.304 +epoch: [253/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:24:49 loss 1.0814 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:24:38 loss 1.0464 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.525, TIME@all 0.303 +epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0706 (1.0631) acc 100.0000 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1212 (1.0723) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.021, TIME@all 0.304 +epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0991 (1.0598) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1128 (1.0712) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.946, TIME@all 0.304 +epoch: [254/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:24:27 loss 1.1010 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1191 (1.0697) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.914, TIME@all 0.304 +epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:24:27 loss 1.0837 (1.0560) acc 100.0000 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:21 loss 1.1003 (1.0622) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.131, TIME@all 0.304 +epoch: [254/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:24:28 loss 1.0613 (1.0545) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0562 (1.0705) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.150, TIME@all 0.304 +epoch: [254/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:24:27 loss 1.0720 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0990 (1.0702) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.946, TIME@all 0.304 +epoch: [254/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0709 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:21 loss 1.1047 (1.0627) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 843.073, TIME@all 0.304 +epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0522 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0478 (1.0635) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.927, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:24:14 loss 1.0631 (1.0603) acc 100.0000 (99.8438) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:08 loss 1.0787 (1.0631) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.014, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:24:14 loss 1.0643 (1.0617) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0924 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.076, TIME@all 0.304 +epoch: [255/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 0:24:14 loss 1.1146 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.1478 (1.0669) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.093, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:24:14 loss 1.0637 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:07 loss 1.0509 (1.0631) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.244, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0622 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0903 (1.0642) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.033, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0902 (1.0611) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0646 (1.0658) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.030, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:24:14 loss 1.0503 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:07 loss 1.0771 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.194, TIME@all 0.304 +epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0756 (1.0621) acc 100.0000 (99.8438) lr 0.002600 +epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0799 (1.0627) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.407, TIME@all 0.304 +epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:01 loss 1.1149 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0679 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.651, TIME@all 0.304 +epoch: [256/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0460 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0548 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.668, TIME@all 0.304 +epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0694 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0625 (1.0653) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.731, TIME@all 0.304 +epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0964 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:23:53 loss 1.0592 (1.0645) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.672, TIME@all 0.304 +epoch: [256/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 0:24:00 loss 1.0632 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0615 (1.0616) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.670, TIME@all 0.304 +epoch: [256/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 0:24:00 loss 1.0604 (1.0501) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:53 loss 1.0453 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.854, TIME@all 0.304 +epoch: [256/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:24:00 loss 1.0820 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0566 (1.0652) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.807, TIME@all 0.304 +epoch: [256/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:24:01 loss 1.0806 (1.0595) acc 100.0000 (99.8438) lr 0.002600 +epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0484 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.822, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:23:41 loss 1.0540 (1.0594) acc 100.0000 (99.8438) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:37 loss 1.0836 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.964, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:23:41 loss 1.1200 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:37 loss 1.0904 (1.0650) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.000, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0997 (1.0625) acc 100.0000 (99.6875) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0490 (1.0695) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.009, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.0953 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0572 (1.0659) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.196, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.1136 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0802 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.998, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.0566 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0636 (1.0672) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.968, TIME@all 0.304 +epoch: [257/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0957 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0677 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.146, TIME@all 0.304 +epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0965 (1.0593) acc 96.8750 (99.6875) lr 0.002600 +epoch: [257/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.1397 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.338, TIME@all 0.304 +epoch: [258/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 0:23:33 loss 1.1860 (1.0625) acc 96.8750 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0890 (1.0713) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.788, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0677 (1.0607) acc 100.0000 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0636 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.849, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.1086 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.1014 (1.0680) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.835, TIME@all 0.304 +epoch: [258/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:23:32 loss 1.1303 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0634 (1.0679) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.871, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:23:32 loss 1.1075 (1.0609) acc 100.0000 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:23:25 loss 1.1182 (1.0702) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 842.016, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0732 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0613 (1.0684) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.160, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0736 (1.0567) acc 100.0000 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 0:23:25 loss 1.0652 (1.0624) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.842, TIME@all 0.304 +epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:23:32 loss 1.0589 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0770 (1.0670) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.971, TIME@all 0.304 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1125 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0465 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.617, TIME@all 0.303 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1359 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0741 (1.0687) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.587, TIME@all 0.303 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.0805 (1.0569) acc 100.0000 (99.8438) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:05 loss 1.0498 (1.0639) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.583, TIME@all 0.303 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:10 loss 1.1501 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:23:05 loss 1.0560 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.783, TIME@all 0.303 +epoch: [259/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:23:10 loss 1.0602 (1.0571) acc 100.0000 (99.8438) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0552 (1.0701) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.614, TIME@all 0.303 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1367 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0523 (1.0692) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.608, TIME@all 0.303 +epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:10 loss 1.1268 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0465 (1.0712) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.752, TIME@all 0.303 +epoch: [259/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.0776 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0586 (1.0648) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.775, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.1067 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0513 (1.0706) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.502, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:23:00 loss 1.0882 (1.0608) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:22:52 loss 1.0473 (1.0666) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.508, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.0799 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0490 (1.0732) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.588, TIME@all 0.303 +epoch: [260/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.1094 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0485 (1.0674) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.515, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:22:59 loss 1.0516 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0492 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.543, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:22:59 loss 1.0580 (1.0611) acc 100.0000 (99.6875) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:22:51 loss 1.1054 (1.0713) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.724, TIME@all 0.303 +epoch: [260/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.0908 (1.0579) acc 100.0000 (99.8438) lr 0.002600 +epoch: [260/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:22:51 loss 1.0482 (1.0720) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.862, TIME@all 0.303 +epoch: [260/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:22:59 loss 1.0716 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:22:51 loss 1.0572 (1.0644) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.658, TIME@all 0.303 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0617 (1.0630) acc 100.0000 (99.8438) lr 0.002600 +epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0810 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 846.116, TIME@all 0.303 +epoch: [261/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0763 (1.0580) acc 100.0000 (99.8438) lr 0.002600 +epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0731 (1.0691) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 846.211, TIME@all 0.303 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0699 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:22:30 loss 1.0580 (1.0661) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 846.267, TIME@all 0.303 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0799 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0705 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 846.093, TIME@all 0.303 +epoch: [261/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0483 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0816 (1.0703) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 846.120, TIME@all 0.303 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0870 (1.0556) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0765 (1.0672) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 846.111, TIME@all 0.303 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0648 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0686 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 846.353, TIME@all 0.302 +epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0732 (1.0530) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:22:30 loss 1.0682 (1.0647) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 846.317, TIME@all 0.302 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0481 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0591 (1.0613) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.959, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0572 (1.0596) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0633 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.999, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0878 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0737 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.008, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0450 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.1449 (1.0646) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.971, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0466 (1.0564) acc 100.0000 (99.8438) lr 0.002600 +epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0705 (1.0682) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.977, TIME@all 0.304 +epoch: [262/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0566 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0618 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.358, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0929 (1.0701) acc 100.0000 (99.6875) lr 0.002600 +epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0756 (1.0741) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.112, TIME@all 0.304 +epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:22:27 loss 1.0685 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:22:22 loss 1.0825 (1.0624) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.168, TIME@all 0.304 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:22:10 loss 1.0522 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:22:04 loss 1.0624 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.504, TIME@all 0.303 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0819 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.0639 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.543, TIME@all 0.303 +epoch: [263/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:22:10 loss 1.0488 (1.0532) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:05 loss 1.0567 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.481, TIME@all 0.304 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0490 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:05 loss 1.0756 (1.0697) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.473, TIME@all 0.304 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:22:09 loss 1.0515 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.1038 (1.0615) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.662, TIME@all 0.303 +epoch: [263/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0893 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.1016 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.541, TIME@all 0.303 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0481 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:22:05 loss 1.0758 (1.0678) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.721, TIME@all 0.303 +epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:22:09 loss 1.0754 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.0852 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.624, TIME@all 0.303 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.011) eta 0:21:59 loss 1.0460 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0653 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.888, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0646 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0493 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.942, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0598 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0995 (1.0692) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.943, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0543 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0466 (1.0678) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.853, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0469 (1.0518) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0694 (1.0681) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.870, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:21:59 loss 1.0536 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:50 loss 1.0488 (1.0653) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.083, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0494 (1.0521) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0571 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.043, TIME@all 0.304 +epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0810 (1.0568) acc 100.0000 (99.8438) lr 0.002600 +epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0614 (1.0684) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.195, TIME@all 0.304 +epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:21:42 loss 1.0571 (1.0620) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:21:33 loss 1.0804 (1.0670) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.212, TIME@all 0.303 +epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0531 (1.0602) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0521 (1.0662) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.143, TIME@all 0.303 +epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0606 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0516 (1.0631) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.194, TIME@all 0.303 +epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0631 (1.0632) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0485 (1.0713) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.195, TIME@all 0.303 +epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0911 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0551 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.311, TIME@all 0.303 +epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0522 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0584 (1.0631) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.125, TIME@all 0.303 +epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0612 (1.0532) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0492 (1.0604) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.359, TIME@all 0.303 +epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:21:42 loss 1.0611 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:21:33 loss 1.0464 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.510, TIME@all 0.303 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:21:28 loss 1.0769 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0854 (1.0723) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.858, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1275 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.1036 (1.0695) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 841.797, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:21:28 loss 1.0653 (1.0601) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.1305 (1.0685) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.758, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1048 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.0991 (1.0664) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 841.771, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:21:28 loss 1.0784 (1.0556) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:22 loss 1.0724 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.968, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1257 (1.0702) acc 100.0000 (99.5312) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.1068 (1.0735) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 841.907, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1555 (1.0650) acc 100.0000 (99.8438) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0872 (1.0684) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 841.996, TIME@all 0.304 +epoch: [266/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:21:28 loss 1.0794 (1.0663) acc 100.0000 (99.6875) lr 0.002600 +epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0614 (1.0695) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 841.788, TIME@all 0.304 +epoch: [267/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.1048 (1.0603) acc 96.8750 (99.6875) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0561 (1.0660) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.263, TIME@all 0.304 +epoch: [267/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0524 (1.0609) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:05 loss 1.0756 (1.0660) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.278, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:11 loss 1.0560 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:05 loss 1.0643 (1.0655) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.341, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0687 (1.0603) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0707 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.239, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0861 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0621 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.252, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:21:11 loss 1.0689 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:05 loss 1.0690 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.454, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:21:11 loss 1.0910 (1.0637) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:05 loss 1.0678 (1.0656) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.425, TIME@all 0.304 +epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0712 (1.0611) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:21:05 loss 1.1059 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.560, TIME@all 0.304 +epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0822 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0737 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 841.976, TIME@all 0.304 +epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0469 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0856 (1.0648) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.989, TIME@all 0.304 +epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0841 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0677 (1.0678) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.019, TIME@all 0.304 +epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.1118 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0758 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.011, TIME@all 0.304 +epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0715 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0567 (1.0631) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 841.991, TIME@all 0.304 +epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0791 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0676 (1.0624) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.379, TIME@all 0.304 +epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0520 (1.0640) acc 100.0000 (99.8438) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0884 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.185, TIME@all 0.304 +epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0549 (1.0596) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0965 (1.0653) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.133, TIME@all 0.304 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:20:40 loss 1.0763 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:20:32 loss 1.1087 (1.0731) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.690, TIME@all 0.303 +epoch: [269/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0659 (1.0578) acc 100.0000 (99.6875) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1135 (1.0717) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.713, TIME@all 0.303 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0629 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1002 (1.0717) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.696, TIME@all 0.303 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0914 (1.0632) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0623 (1.0706) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.758, TIME@all 0.303 +epoch: [269/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.1278 (1.0603) acc 96.8750 (99.6875) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0662 (1.0697) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.679, TIME@all 0.303 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:20:40 loss 1.0484 (1.0575) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1032 (1.0684) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.886, TIME@all 0.303 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 0:20:40 loss 1.0689 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0780 (1.0675) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.000, TIME@all 0.303 +epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:20:40 loss 1.0495 (1.0580) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1145 (1.0671) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.833, TIME@all 0.303 +epoch: [270/350][20/50] time 0.297 (0.304) data 0.000 (0.011) eta 0:20:23 loss 1.0846 (1.0626) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0518 (1.0631) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.460, TIME@all 0.304 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0571 (1.0564) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0501 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.539, TIME@all 0.303 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:22 loss 1.0551 (1.0572) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0668 (1.0646) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.490, TIME@all 0.304 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:23 loss 1.0459 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:20:19 loss 1.0570 (1.0679) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.495, TIME@all 0.303 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:22 loss 1.0465 (1.0552) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:18 loss 1.0542 (1.0616) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.633, TIME@all 0.303 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0484 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0650 (1.0677) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.492, TIME@all 0.304 +epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:22 loss 1.0578 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:20:18 loss 1.0620 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.686, TIME@all 0.303 +epoch: [270/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0675 (1.0501) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0962 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.827, TIME@all 0.303 +epoch: [271/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1264 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0698 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.278, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1049 (1.0637) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0662 (1.0753) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.311, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.0762 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.1155 (1.0746) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.330, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1351 (1.0642) acc 96.8750 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.1669 (1.0705) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 842.264, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:20:13 loss 1.1678 (1.0715) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:20:06 loss 1.0944 (1.0753) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.494, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.1332 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0688 (1.0741) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.281, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:20:13 loss 1.1086 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:20:06 loss 1.0906 (1.0757) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.429, TIME@all 0.304 +epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.0917 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0969 (1.0764) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.658, TIME@all 0.304 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0614 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0596 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.817, TIME@all 0.303 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0492 (1.0515) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0809 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.839, TIME@all 0.303 +epoch: [272/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0477 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0584 (1.0678) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.892, TIME@all 0.303 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:19:56 loss 1.0579 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:19:48 loss 1.1278 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.006, TIME@all 0.303 +epoch: [272/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:19:56 loss 1.0674 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0495 (1.0640) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.019, TIME@all 0.303 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0429 (1.0511) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0637 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.822, TIME@all 0.303 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0633 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0665 (1.0659) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.972, TIME@all 0.303 +epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0593 (1.0493) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0537 (1.0622) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.803, TIME@all 0.303 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0812 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0631 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.135, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:40 loss 1.0699 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:19:34 loss 1.0633 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.052, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0731 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:19:34 loss 1.0563 (1.0632) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.160, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0629 (1.0549) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0552 (1.0641) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.079, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:19:40 loss 1.0877 (1.0626) acc 100.0000 (99.6875) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:33 loss 1.0776 (1.0635) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.265, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.1475 (1.0679) acc 96.8750 (99.5312) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0681 (1.0674) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.084, TIME@all 0.304 +epoch: [273/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:19:40 loss 1.0746 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:33 loss 1.0526 (1.0698) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.252, TIME@all 0.304 +epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:41 loss 1.0751 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0627 (1.0588) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.260, TIME@all 0.304 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0620 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:16 loss 1.0541 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.006, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.1743 (1.0624) acc 96.8750 (99.8438) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0619 (1.0692) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.884, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 0:19:25 loss 1.0525 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0576 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.889, TIME@all 0.303 +epoch: [274/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0501 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:16 loss 1.0554 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.266, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0649 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0730 (1.0624) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.895, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:25 loss 1.0498 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:16 loss 1.0852 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.112, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0641 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0505 (1.0612) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.906, TIME@all 0.303 +epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:25 loss 1.0622 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:19:16 loss 1.0938 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.053, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0961 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:19:01 loss 1.0700 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.735, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.013) eta 0:19:05 loss 1.0860 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:02 loss 1.0572 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.680, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0836 (1.0520) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:19:02 loss 1.1468 (1.0645) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.658, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.011) eta 0:19:05 loss 1.0763 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:19:02 loss 1.0665 (1.0683) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.662, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.014) eta 0:19:05 loss 1.0668 (1.0599) acc 100.0000 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:01 loss 1.0502 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.843, TIME@all 0.303 +epoch: [275/350][20/50] time 0.297 (0.303) data 0.001 (0.012) eta 0:19:05 loss 1.1076 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:19:01 loss 1.1221 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.658, TIME@all 0.303 +epoch: [275/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0714 (1.0631) acc 100.0000 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:19:01 loss 1.0900 (1.0698) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.019, TIME@all 0.303 +epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.013) eta 0:19:05 loss 1.0585 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:01 loss 1.1035 (1.0742) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.780, TIME@all 0.303 +epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0670 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0632 (1.0715) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.284, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0757 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0584 (1.0701) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.208, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0678 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.1087 (1.0689) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.246, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0778 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0803 (1.0694) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.277, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.1130 (1.0629) acc 96.8750 (99.8438) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.2308 (1.0750) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 843.466, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.0488 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0567 (1.0705) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.415, TIME@all 0.304 +epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:18:55 loss 1.0651 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0903 (1.0672) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.252, TIME@all 0.304 +epoch: [276/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.0478 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0819 (1.0697) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.591, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0512 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.1427 (1.0732) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 844.865, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0702 (1.0613) acc 100.0000 (99.8438) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.0508 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.903, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0585 (1.0581) acc 100.0000 (99.8438) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:18:29 loss 1.0751 (1.0658) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.021, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0540 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:18:29 loss 1.0959 (1.0664) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 845.060, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0622 (1.0611) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.1115 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.927, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0681 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:18:29 loss 1.1070 (1.0657) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 844.845, TIME@all 0.303 +epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0711 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.006) eta 0:18:29 loss 1.0635 (1.0669) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.873, TIME@all 0.303 +epoch: [277/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0669 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.0856 (1.0698) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 845.275, TIME@all 0.303 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0979 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0538 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.959, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:18:25 loss 1.0705 (1.0520) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:18:18 loss 1.0525 (1.0660) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.871, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:18:25 loss 1.0723 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:18:18 loss 1.0634 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.889, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0879 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0549 (1.0646) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.897, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0757 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0818 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.905, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:18:25 loss 1.0746 (1.0549) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0585 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.083, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.1031 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0545 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.028, TIME@all 0.304 +epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0924 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0620 (1.0662) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.244, TIME@all 0.304 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0683 (1.0519) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0861 (1.0631) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.834, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0710 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0912 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.891, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0641 (1.0512) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1315 (1.0630) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.917, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:18:07 loss 1.0997 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:18:01 loss 1.0756 (1.0665) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.040, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0874 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0880 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.827, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0630 (1.0519) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1170 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.839, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0904 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1431 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.982, TIME@all 0.303 +epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0675 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0764 (1.0601) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.053, TIME@all 0.303 +epoch: [280/350][20/50] time 0.312 (0.307) data 0.001 (0.012) eta 0:18:02 loss 1.0533 (1.0573) acc 100.0000 (99.8438) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:17:50 loss 1.0461 (1.0629) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 840.540, TIME@all 0.305 +epoch: [280/350][20/50] time 0.297 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0650 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0533 (1.0646) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 839.562, TIME@all 0.305 +epoch: [280/350][20/50] time 0.295 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0594 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:17:52 loss 1.0547 (1.0610) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.563, TIME@all 0.305 +epoch: [280/350][20/50] time 0.308 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.1720 (1.0624) acc 96.8750 (99.8438) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:52 loss 1.0489 (1.0695) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.499, TIME@all 0.305 +epoch: [280/350][20/50] time 0.309 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0662 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:52 loss 1.0575 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 839.487, TIME@all 0.305 +epoch: [280/350][20/50] time 0.308 (0.307) data 0.000 (0.014) eta 0:18:02 loss 1.0684 (1.0595) acc 100.0000 (99.8438) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0538 (1.0701) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 839.673, TIME@all 0.305 +epoch: [280/350][20/50] time 0.309 (0.307) data 0.000 (0.014) eta 0:18:02 loss 1.0594 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0601 (1.0697) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 839.623, TIME@all 0.305 +epoch: [280/350][20/50] time 0.306 (0.308) data 0.000 (0.013) eta 0:18:06 loss 1.0705 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:53 loss 1.0543 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 838.894, TIME@all 0.305 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:17:37 loss 1.0597 (1.0586) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0692 (1.0652) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.894, TIME@all 0.303 +epoch: [281/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0612 (1.0597) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0702 (1.0693) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.946, TIME@all 0.303 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.1250 (1.0636) acc 96.8750 (99.5312) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0493 (1.0690) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.992, TIME@all 0.303 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:17:37 loss 1.0574 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:17:30 loss 1.0405 (1.0675) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.120, TIME@all 0.303 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0519 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0672 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.053, TIME@all 0.303 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.1104 (1.0607) acc 96.8750 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0530 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.900, TIME@all 0.303 +epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0578 (1.0610) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0766 (1.0633) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.929, TIME@all 0.303 +epoch: [281/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0561 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [281/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0460 (1.0644) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.288, TIME@all 0.303 +epoch: [282/350][20/50] time 0.302 (0.303) data 0.001 (0.012) eta 0:17:20 loss 1.1048 (1.0598) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0745 (1.0647) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.669, TIME@all 0.304 +epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:17:20 loss 1.0925 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0690 (1.0648) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.679, TIME@all 0.304 +epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:17:20 loss 1.0776 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0476 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.748, TIME@all 0.304 +epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0966 (1.0564) acc 100.0000 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0523 (1.0660) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 842.883, TIME@all 0.304 +epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0591 (1.0523) acc 100.0000 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0458 (1.0612) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.675, TIME@all 0.304 +epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.1471 (1.0575) acc 96.8750 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0600 (1.0621) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.980, TIME@all 0.304 +epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0932 (1.0593) acc 100.0000 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0532 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.694, TIME@all 0.304 +epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.1286 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0808 (1.0665) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.821, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 0:17:11 loss 1.0579 (1.0543) acc 100.0000 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.1423 (1.0696) acc 96.8750 (99.6875) lr 0.002600 +FPS@all 843.036, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0686 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0746 (1.0681) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.129, TIME@all 0.304 +epoch: [283/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0552 (1.0503) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0790 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.070, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 0:17:11 loss 1.0480 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0966 (1.0662) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.080, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:17:11 loss 1.1076 (1.0576) acc 96.8750 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:17:02 loss 1.0765 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.280, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0530 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0583 (1.0672) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.096, TIME@all 0.304 +epoch: [283/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:17:12 loss 1.0511 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:17:02 loss 1.0594 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.397, TIME@all 0.304 +epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0531 (1.0569) acc 100.0000 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:17:02 loss 1.0600 (1.0670) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.232, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1720 (1.0671) acc 96.8750 (99.6875) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2800 (1.0765) acc 93.7500 (99.5312) lr 0.002600 +FPS@all 843.314, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1454 (1.0575) acc 96.8750 (99.8438) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1541 (1.0726) acc 96.8750 (99.6094) lr 0.002600 +FPS@all 843.373, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.0751 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1162 (1.0679) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.340, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1062 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2198 (1.0676) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.348, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1267 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2419 (1.0765) acc 96.8750 (99.6094) lr 0.002600 +FPS@all 843.323, TIME@all 0.304 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:16:54 loss 1.0534 (1.0565) acc 100.0000 (99.8438) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:16:47 loss 1.2155 (1.0670) acc 93.7500 (99.6875) lr 0.002600 +FPS@all 843.530, TIME@all 0.303 +epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:16:55 loss 1.1203 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1152 (1.0699) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.474, TIME@all 0.304 +epoch: [284/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:16:54 loss 1.0785 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2171 (1.0727) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 843.557, TIME@all 0.303 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:16:40 loss 1.2068 (1.0705) acc 96.8750 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:16:32 loss 1.0658 (1.0721) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.262, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:16:40 loss 1.1491 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:16:32 loss 1.0466 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.137, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.1622 (1.0627) acc 100.0000 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0568 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.199, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0785 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0520 (1.0635) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.179, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0962 (1.0617) acc 100.0000 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0559 (1.0697) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.185, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.1104 (1.0635) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0450 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.378, TIME@all 0.304 +epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0891 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0778 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.318, TIME@all 0.304 +epoch: [285/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0589 (1.0600) acc 100.0000 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0484 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.591, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0980 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0531 (1.0631) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.188, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0686 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0481 (1.0616) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.265, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:16:22 loss 1.0631 (1.0501) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.001 (0.007) eta 0:16:16 loss 1.0571 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.417, TIME@all 0.304 +epoch: [286/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:16:23 loss 1.1103 (1.0581) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.001 (0.007) eta 0:16:17 loss 1.0535 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.194, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.1400 (1.0573) acc 96.8750 (99.8438) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0612 (1.0677) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.223, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.1348 (1.0567) acc 100.0000 (99.8438) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0547 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.235, TIME@all 0.304 +epoch: [286/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0964 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0530 (1.0634) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.522, TIME@all 0.304 +epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:16:22 loss 1.0648 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:16 loss 1.0622 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.362, TIME@all 0.304 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.011) eta 0:16:05 loss 1.0551 (1.0577) acc 100.0000 (99.8438) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0667 (1.0677) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.969, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0588 (1.0518) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0522 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.002, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0813 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0642 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.037, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.1032 (1.0617) acc 96.8750 (99.8438) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0709 (1.0695) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.181, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0818 (1.0545) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0715 (1.0671) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.015, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0627 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.1223 (1.0679) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 843.986, TIME@all 0.303 +epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:16:05 loss 1.0684 (1.0578) acc 100.0000 (99.8438) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0523 (1.0666) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.125, TIME@all 0.303 +epoch: [287/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0539 (1.0520) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0563 (1.0606) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.293, TIME@all 0.303 +epoch: [288/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:15:51 loss 1.0736 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:15:44 loss 1.0518 (1.0669) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.549, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:15:51 loss 1.0477 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:15:44 loss 1.0719 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.480, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.1112 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0410 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.512, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:15:51 loss 1.0542 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:15:44 loss 1.0574 (1.0623) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.726, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0803 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0481 (1.0633) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.535, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:15:51 loss 1.0808 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0443 (1.0644) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.502, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0653 (1.0609) acc 100.0000 (99.8438) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0555 (1.0664) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.841, TIME@all 0.303 +epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0884 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0457 (1.0660) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.680, TIME@all 0.303 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:15:36 loss 1.1834 (1.0637) acc 96.8750 (99.8438) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.0840 (1.0735) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 842.896, TIME@all 0.304 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:15:36 loss 1.0858 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.1295 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.961, TIME@all 0.304 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0950 (1.0606) acc 100.0000 (99.8438) lr 0.002600 +epoch: [289/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:15:30 loss 1.0636 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.899, TIME@all 0.304 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0959 (1.0568) acc 96.8750 (99.8438) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.0621 (1.0660) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.943, TIME@all 0.304 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0726 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.0860 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.916, TIME@all 0.304 +epoch: [289/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:15:35 loss 1.0827 (1.0518) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.1400 (1.0691) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 843.115, TIME@all 0.304 +epoch: [289/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.1103 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.0478 (1.0650) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.258, TIME@all 0.304 +epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:35 loss 1.1344 (1.0572) acc 96.8750 (99.8438) lr 0.002600 +epoch: [289/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.1097 (1.0656) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.094, TIME@all 0.304 +epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0661 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0539 (1.0649) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.182, TIME@all 0.304 +epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0466 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0443 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.221, TIME@all 0.304 +epoch: [290/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0681 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0660 (1.0647) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.206, TIME@all 0.304 +epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0706 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:15 loss 1.0493 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.202, TIME@all 0.304 +epoch: [290/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0420 (1.0532) acc 100.0000 (99.8438) lr 0.002600 +epoch: [290/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0488 (1.0671) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.562, TIME@all 0.303 +epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0651 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:14 loss 1.0478 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.371, TIME@all 0.304 +epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0685 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0675 (1.0707) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.176, TIME@all 0.304 +epoch: [290/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0498 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:14 loss 1.0552 (1.0687) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.355, TIME@all 0.304 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.0687 (1.0522) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.1356 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.735, TIME@all 0.303 +epoch: [291/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.1037 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.1135 (1.0667) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.830, TIME@all 0.303 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.0501 (1.0567) acc 100.0000 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.0736 (1.0657) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.746, TIME@all 0.303 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.0569 (1.0584) acc 100.0000 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.1279 (1.0713) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.777, TIME@all 0.303 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.1646 (1.0621) acc 96.8750 (99.6875) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0843 (1.0733) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 843.762, TIME@all 0.303 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:15:06 loss 1.0796 (1.0667) acc 100.0000 (99.6875) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0588 (1.0677) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.949, TIME@all 0.303 +epoch: [291/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.1336 (1.0576) acc 96.8750 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0545 (1.0617) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.087, TIME@all 0.303 +epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.0547 (1.0515) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0941 (1.0653) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.900, TIME@all 0.303 +epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.1118 (1.0599) acc 100.0000 (99.8438) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:44 loss 1.0494 (1.0686) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.326, TIME@all 0.304 +epoch: [292/350][20/50] time 0.304 (0.304) data 0.001 (0.012) eta 0:14:50 loss 1.0617 (1.0514) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0601 (1.0659) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.253, TIME@all 0.304 +epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0708 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0668 (1.0718) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.311, TIME@all 0.304 +epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0704 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0709 (1.0728) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.477, TIME@all 0.304 +epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0619 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0592 (1.0717) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.275, TIME@all 0.304 +epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0748 (1.0507) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:44 loss 1.0420 (1.0654) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.284, TIME@all 0.304 +epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0749 (1.0578) acc 100.0000 (99.8438) lr 0.002600 +epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0517 (1.0700) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.428, TIME@all 0.304 +epoch: [292/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0821 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0470 (1.0689) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.635, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0451 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0497 (1.0643) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.265, TIME@all 0.303 +epoch: [293/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0656 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0867 (1.0694) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.103, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.1205 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0840 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 845.207, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0546 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0547 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.146, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.1232 (1.0598) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0554 (1.0639) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 845.168, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0474 (1.0603) acc 100.0000 (99.8438) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.1093 (1.0679) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 845.358, TIME@all 0.303 +epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0545 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0809 (1.0694) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.291, TIME@all 0.303 +epoch: [293/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0629 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0621 (1.0686) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 845.392, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:14:17 loss 1.0395 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:14:14 loss 1.0578 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.909, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0541 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0542 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 843.777, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:17 loss 1.0671 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0737 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.789, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.014) eta 0:14:17 loss 1.0531 (1.0523) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0612 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.979, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0566 (1.0512) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0678 (1.0600) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.834, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0731 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0563 (1.0653) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.787, TIME@all 0.303 +epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:17 loss 1.0514 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0775 (1.0661) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.132, TIME@all 0.303 +epoch: [294/350][20/50] time 0.303 (0.303) data 0.001 (0.013) eta 0:14:17 loss 1.0552 (1.0546) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0593 (1.0627) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.945, TIME@all 0.303 +epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:14:01 loss 1.0501 (1.0563) acc 100.0000 (99.8438) lr 0.002600 +epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:13:57 loss 1.0683 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.721, TIME@all 0.303 +epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:14:01 loss 1.0579 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:13:57 loss 1.1048 (1.0636) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.665, TIME@all 0.303 +epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0479 (1.0588) acc 100.0000 (99.6875) lr 0.002600 +epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0863 (1.0691) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.695, TIME@all 0.303 +epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0584 (1.0513) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0929 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.676, TIME@all 0.303 +epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0456 (1.0518) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1197 (1.0620) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.859, TIME@all 0.303 +epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0485 (1.0575) acc 100.0000 (99.8438) lr 0.002600 +epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1007 (1.0711) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.667, TIME@all 0.303 +epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0518 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0784 (1.0656) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.821, TIME@all 0.303 +epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0674 (1.0615) acc 100.0000 (99.8438) lr 0.002600 +epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1069 (1.0724) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 845.137, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1634 (1.0658) acc 96.8750 (99.5312) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0602 (1.0664) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.215, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1086 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0413 (1.0665) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.151, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:13:48 loss 1.2177 (1.0616) acc 96.8750 (99.6875) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:13:43 loss 1.1168 (1.0727) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 844.318, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1431 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0646 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.130, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:48 loss 1.1645 (1.0683) acc 96.8750 (99.6875) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:13:43 loss 1.0715 (1.0737) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 844.379, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1463 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0698 (1.0731) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.189, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1283 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0569 (1.0633) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.179, TIME@all 0.303 +epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1528 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0539 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.510, TIME@all 0.303 +epoch: [297/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0772 (1.0523) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.1012 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.001, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0672 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0883 (1.0670) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 843.004, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0754 (1.0550) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0877 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.033, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0859 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.1186 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.975, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:34 loss 1.0593 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:13:29 loss 1.0604 (1.0689) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 843.172, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0704 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0894 (1.0633) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 843.121, TIME@all 0.304 +epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0606 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.0660 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.964, TIME@all 0.304 +epoch: [297/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0706 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0829 (1.0661) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 843.313, TIME@all 0.304 +epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0506 (1.0529) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.1675 (1.0723) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.243, TIME@all 0.303 +epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.1373 (1.0630) acc 96.8750 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.0736 (1.0727) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.249, TIME@all 0.303 +epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0638 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1242 (1.0705) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.277, TIME@all 0.303 +epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0563 (1.0596) acc 100.0000 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.1070 (1.0652) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.273, TIME@all 0.303 +epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0493 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1960 (1.0693) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 844.460, TIME@all 0.303 +epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0650 (1.0590) acc 100.0000 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:13 loss 1.0687 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.607, TIME@all 0.303 +epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0585 (1.0627) acc 100.0000 (99.6875) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1180 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.263, TIME@all 0.303 +epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0559 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1101 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.420, TIME@all 0.303 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.011) eta 0:13:04 loss 1.0741 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.1183 (1.0694) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.189, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.1197 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.0744 (1.0753) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.278, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.1092 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.0690 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 842.231, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:04 loss 1.1008 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:12:59 loss 1.0770 (1.0682) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.204, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:03 loss 1.0986 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 0:12:59 loss 1.0978 (1.0693) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 842.418, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.0732 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.309 (0.304) data 0.001 (0.006) eta 0:12:59 loss 1.0619 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 842.187, TIME@all 0.304 +epoch: [299/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:03 loss 1.0703 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.1044 (1.0729) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 842.563, TIME@all 0.304 +epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:04 loss 1.0924 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:12:59 loss 1.0985 (1.0738) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 842.367, TIME@all 0.304 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0527 (1.0507) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0427 (1.0642) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.477, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:12:48 loss 1.0526 (1.0532) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:12:43 loss 1.0642 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.408, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:12:48 loss 1.0585 (1.0581) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:12:43 loss 1.0474 (1.0692) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.487, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:12:48 loss 1.0780 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0637 (1.0670) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 844.580, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0501 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0477 (1.0695) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.406, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0724 (1.0539) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0564 (1.0699) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 844.408, TIME@all 0.303 +epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0697 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0597 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 844.557, TIME@all 0.303 +epoch: [300/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0555 (1.0539) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0661 (1.0719) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 844.767, TIME@all 0.303 +epoch: [301/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0613 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0688 (1.0622) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.063, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0478 (1.0518) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0621 (1.0612) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 845.087, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0659 (1.0536) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0626 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.983, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:12:32 loss 1.0816 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.306 (0.303) data 0.001 (0.007) eta 0:12:26 loss 1.0588 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 845.201, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0474 (1.0500) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0862 (1.0618) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.037, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0599 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.305 (0.303) data 0.001 (0.006) eta 0:12:26 loss 1.0718 (1.0663) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.026, TIME@all 0.303 +epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:12:32 loss 1.0671 (1.0585) acc 100.0000 (99.8438) lr 0.000260 +epoch: [301/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:12:26 loss 1.0885 (1.0695) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 845.159, TIME@all 0.303 +epoch: [301/350][20/50] time 0.301 (0.303) data 0.001 (0.012) eta 0:12:32 loss 1.0705 (1.0556) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.305 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0630 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 845.315, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.011) eta 0:12:17 loss 1.0545 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.1234 (1.0642) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 843.593, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0607 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0692 (1.0625) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.664, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0515 (1.0529) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:12:12 loss 1.0702 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.804, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0560 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0612 (1.0671) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.606, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0507 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0929 (1.0646) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.612, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0614 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0843 (1.0630) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.629, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0459 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0934 (1.0620) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.759, TIME@all 0.303 +epoch: [302/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:12:17 loss 1.0646 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0833 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.748, TIME@all 0.303 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0552 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1023 (1.0691) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.355, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.011) eta 0:12:03 loss 1.0817 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0673 (1.0661) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.273, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.1474 (1.0563) acc 96.8750 (99.6875) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0567 (1.0650) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 842.366, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0795 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0559 (1.0672) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.299, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0639 (1.0606) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1074 (1.0709) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.327, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0652 (1.0569) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0723 (1.0649) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.676, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:12:03 loss 1.0798 (1.0552) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:11:58 loss 1.0554 (1.0693) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.481, TIME@all 0.304 +epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:12:03 loss 1.0756 (1.0622) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1654 (1.0738) acc 96.8750 (99.7656) lr 0.000260 +FPS@all 842.447, TIME@all 0.304 +epoch: [304/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0579 (1.0548) acc 100.0000 (99.8438) lr 0.000260 +epoch: [304/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:11:41 loss 1.1258 (1.0630) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.293, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0555 (1.0578) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0689 (1.0661) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.237, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:11:49 loss 1.0682 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:41 loss 1.0641 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.179, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:11:48 loss 1.0505 (1.0595) acc 100.0000 (99.8438) lr 0.000260 +epoch: [304/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:11:40 loss 1.1018 (1.0694) acc 96.8750 (99.6875) lr 0.000260 +FPS@all 845.371, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0590 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.301 (0.303) data 0.001 (0.007) eta 0:11:41 loss 1.1090 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 845.564, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0491 (1.0556) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0495 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.173, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0767 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0829 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.204, TIME@all 0.303 +epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0511 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:11:40 loss 1.0836 (1.0622) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.329, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.011) eta 0:11:32 loss 1.0647 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0528 (1.0639) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.663, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:11:32 loss 1.0748 (1.0575) acc 100.0000 (99.8438) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0769 (1.0652) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.707, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:11:32 loss 1.1233 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0491 (1.0658) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.744, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0719 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0806 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.694, TIME@all 0.303 +epoch: [305/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:32 loss 1.1035 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0646 (1.0647) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.996, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0791 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0515 (1.0709) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.832, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:11:32 loss 1.1262 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:26 loss 1.0715 (1.0681) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.890, TIME@all 0.303 +epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0658 (1.0550) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0539 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.689, TIME@all 0.303 +epoch: [306/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:11:17 loss 1.1035 (1.0589) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1163 (1.0724) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.654, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:17 loss 1.1553 (1.0595) acc 96.8750 (99.8438) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1108 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.686, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:17 loss 1.1215 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1183 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.707, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:17 loss 1.1302 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:11 loss 1.0618 (1.0696) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.859, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:11:17 loss 1.0814 (1.0606) acc 100.0000 (99.8438) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:11 loss 1.0912 (1.0700) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.653, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.001 (0.012) eta 0:11:17 loss 1.1069 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1218 (1.0674) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.636, TIME@all 0.303 +epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:17 loss 1.1202 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.0693 (1.0649) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.772, TIME@all 0.303 +epoch: [306/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 0:11:17 loss 1.0792 (1.0570) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.0767 (1.0636) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.038, TIME@all 0.303 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.1065 (1.0654) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0494 (1.0676) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.331, TIME@all 0.304 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.0613 (1.0666) acc 100.0000 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0532 (1.0717) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.388, TIME@all 0.304 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0517 (1.0605) acc 100.0000 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0524 (1.0745) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 843.375, TIME@all 0.304 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0660 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0617 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.338, TIME@all 0.304 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.1055 (1.0618) acc 96.8750 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0463 (1.0715) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.513, TIME@all 0.303 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0868 (1.0643) acc 100.0000 (99.6875) lr 0.000260 +epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0612 (1.0657) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.485, TIME@all 0.304 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.0701 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0967 (1.0703) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.715, TIME@all 0.303 +epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0865 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:10:56 loss 1.0514 (1.0697) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.338, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:10:48 loss 1.0724 (1.0556) acc 100.0000 (99.8438) lr 0.000260 +epoch: [308/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0534 (1.0604) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.454, TIME@all 0.304 +epoch: [308/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:10:48 loss 1.0874 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0491 (1.0650) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.396, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:10:48 loss 1.0871 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0535 (1.0628) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.427, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:10:48 loss 1.0875 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0440 (1.0613) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.386, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.0797 (1.0572) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0484 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.418, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.1035 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0409 (1.0667) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.604, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.1178 (1.0629) acc 96.8750 (99.8438) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0452 (1.0691) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.768, TIME@all 0.304 +epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.0775 (1.0499) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0857 (1.0670) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.562, TIME@all 0.304 +epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0697 (1.0597) acc 100.0000 (99.8438) lr 0.000260 +epoch: [309/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.1301 (1.0670) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 843.768, TIME@all 0.303 +epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0584 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:10:25 loss 1.0552 (1.0639) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.826, TIME@all 0.303 +epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0471 (1.0584) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0546 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.799, TIME@all 0.303 +epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.0664 (1.0650) acc 100.0000 (99.6875) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0533 (1.0679) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.997, TIME@all 0.303 +epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0701 (1.0587) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0719 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.792, TIME@all 0.303 +epoch: [309/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0475 (1.0573) acc 100.0000 (99.8438) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0943 (1.0649) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.177, TIME@all 0.303 +epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.0660 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0610 (1.0621) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.947, TIME@all 0.303 +epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.1896 (1.0657) acc 96.8750 (99.8438) lr 0.000260 +epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0939 (1.0695) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.799, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0579 (1.0514) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0590 (1.0615) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.754, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0616 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.299 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0726 (1.0703) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.779, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0741 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.1570 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.910, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0885 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.300 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0719 (1.0646) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.777, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0433 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.001 (0.006) eta 0:10:09 loss 1.0671 (1.0659) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.772, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0764 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:10:09 loss 1.0769 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.971, TIME@all 0.303 +epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0544 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:10:09 loss 1.0598 (1.0656) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 844.928, TIME@all 0.303 +epoch: [310/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0544 (1.0498) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0494 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.145, TIME@all 0.303 +epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0659 (1.0529) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0601 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.895, TIME@all 0.303 +epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0603 (1.0511) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0703 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.930, TIME@all 0.303 +epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:10:01 loss 1.0614 (1.0500) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.006) eta 0:09:54 loss 1.0457 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.880, TIME@all 0.303 +epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0711 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0582 (1.0593) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.986, TIME@all 0.303 +epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0629 (1.0493) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0653 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.909, TIME@all 0.303 +epoch: [311/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:10:01 loss 1.0621 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0477 (1.0630) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 845.086, TIME@all 0.303 +epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0747 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0570 (1.0631) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.039, TIME@all 0.303 +epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0526 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0625 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 845.226, TIME@all 0.303 +epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0605 (1.0649) acc 100.0000 (99.8438) lr 0.000260 +epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.0668 (1.0677) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.749, TIME@all 0.303 +epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:09:45 loss 1.0585 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:09:40 loss 1.0531 (1.0649) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.651, TIME@all 0.303 +epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0734 (1.0605) acc 100.0000 (99.8438) lr 0.000260 +epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1035 (1.0662) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.637, TIME@all 0.303 +epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 0:09:45 loss 1.0490 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1028 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.859, TIME@all 0.303 +epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0521 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1479 (1.0644) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 844.062, TIME@all 0.303 +epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0524 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1189 (1.0651) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.674, TIME@all 0.303 +epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0502 (1.0618) acc 100.0000 (99.6875) lr 0.000260 +epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1560 (1.0675) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.833, TIME@all 0.303 +epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0651 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.0988 (1.0624) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.699, TIME@all 0.303 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:09:31 loss 1.0488 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0566 (1.0684) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.485, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0550 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0679 (1.0642) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.533, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0600 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0884 (1.0685) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.580, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0687 (1.0632) acc 100.0000 (99.8438) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0579 (1.0685) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.485, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:09:31 loss 1.1140 (1.0663) acc 96.8750 (99.6875) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0846 (1.0742) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 842.506, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:09:31 loss 1.0889 (1.0546) acc 100.0000 (99.8438) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:09:25 loss 1.0824 (1.0662) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.696, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:09:31 loss 1.0480 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:09:25 loss 1.0472 (1.0658) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.647, TIME@all 0.304 +epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0686 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0690 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.839, TIME@all 0.304 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0552 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.0599 (1.0618) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.943, TIME@all 0.303 +epoch: [314/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0494 (1.0533) acc 100.0000 (99.8438) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.0736 (1.0605) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.947, TIME@all 0.303 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0923 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.1173 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.995, TIME@all 0.303 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0633 (1.0504) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0672 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.941, TIME@all 0.303 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0554 (1.0549) acc 100.0000 (99.8438) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.1298 (1.0723) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 843.961, TIME@all 0.303 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0446 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0604 (1.0651) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.115, TIME@all 0.303 +epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0595 (1.0510) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0693 (1.0611) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.079, TIME@all 0.303 +epoch: [314/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0495 (1.0494) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0498 (1.0642) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.202, TIME@all 0.303 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:09:01 loss 1.0857 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1019 (1.0668) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.405, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:09:01 loss 1.1000 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1180 (1.0671) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.406, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0835 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0897 (1.0656) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.390, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0636 (1.0593) acc 100.0000 (99.8438) lr 0.000260 +epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0980 (1.0643) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.459, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:09:01 loss 1.0724 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:55 loss 1.0716 (1.0691) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.550, TIME@all 0.304 +epoch: [315/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0779 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0567 (1.0645) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.678, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.1024 (1.0581) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0769 (1.0648) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.546, TIME@all 0.304 +epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0836 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1057 (1.0655) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.397, TIME@all 0.304 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:08:47 loss 1.0600 (1.0558) acc 100.0000 (99.8438) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0691 (1.0668) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 844.045, TIME@all 0.303 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:08:47 loss 1.0526 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0479 (1.0616) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.083, TIME@all 0.303 +epoch: [316/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:08:47 loss 1.0455 (1.0568) acc 100.0000 (99.8438) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0438 (1.0629) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.058, TIME@all 0.303 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:46 loss 1.0539 (1.0505) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:08:39 loss 1.0596 (1.0621) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.241, TIME@all 0.303 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:08:47 loss 1.0546 (1.0516) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0551 (1.0619) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.029, TIME@all 0.303 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:46 loss 1.0428 (1.0526) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0725 (1.0634) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.188, TIME@all 0.303 +epoch: [316/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:08:46 loss 1.0473 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0807 (1.0635) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.475, TIME@all 0.303 +epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:08:46 loss 1.0510 (1.0511) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0587 (1.0636) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.039, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:08:29 loss 1.0677 (1.0600) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:24 loss 1.0456 (1.0649) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.523, TIME@all 0.303 +epoch: [317/350][20/50] time 0.300 (0.303) data 0.001 (0.013) eta 0:08:29 loss 1.0457 (1.0645) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0897 (1.0719) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.584, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0451 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0647 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.542, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0547 (1.0558) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0757 (1.0666) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.585, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0528 (1.0637) acc 100.0000 (99.6875) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0925 (1.0737) acc 96.8750 (99.6875) lr 0.000260 +FPS@all 843.551, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:29 loss 1.0488 (1.0599) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0697 (1.0699) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.702, TIME@all 0.303 +epoch: [317/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0571 (1.0626) acc 100.0000 (99.8438) lr 0.000260 +epoch: [317/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0593 (1.0670) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.928, TIME@all 0.303 +epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:29 loss 1.0660 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.1176 (1.0682) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.723, TIME@all 0.303 +epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:08:14 loss 1.0630 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0581 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.926, TIME@all 0.303 +epoch: [318/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:08:14 loss 1.0654 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0669 (1.0658) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.965, TIME@all 0.303 +epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.0662 (1.0602) acc 100.0000 (99.8438) lr 0.000260 +epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0562 (1.0668) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.928, TIME@all 0.303 +epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.0802 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:08:09 loss 1.0619 (1.0632) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.920, TIME@all 0.303 +epoch: [318/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:14 loss 1.0623 (1.0521) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0575 (1.0615) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.088, TIME@all 0.303 +epoch: [318/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:08:14 loss 1.0621 (1.0584) acc 100.0000 (99.8438) lr 0.000260 +epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0472 (1.0664) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.962, TIME@all 0.303 +epoch: [318/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:08:14 loss 1.0682 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0487 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.036, TIME@all 0.303 +epoch: [318/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.1093 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0613 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.224, TIME@all 0.303 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0653 (1.0554) acc 100.0000 (99.8438) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0478 (1.0607) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.400, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.1132 (1.0622) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0458 (1.0656) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.380, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0745 (1.0577) acc 100.0000 (99.8438) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0461 (1.0570) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.465, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0570 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0536 (1.0636) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.379, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0783 (1.0594) acc 100.0000 (99.8438) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:07:54 loss 1.0458 (1.0612) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.548, TIME@all 0.304 +epoch: [319/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.1104 (1.0559) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:07:54 loss 1.0987 (1.0643) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.627, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0441 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:07:54 loss 1.0558 (1.0598) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.391, TIME@all 0.304 +epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0625 (1.0570) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:07:54 loss 1.1173 (1.0615) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.565, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0734 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0504 (1.0615) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.822, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0989 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0476 (1.0637) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.879, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0758 (1.0607) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0511 (1.0700) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.873, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0749 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0874 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.909, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1238 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:07:39 loss 1.0557 (1.0631) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.858, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:07:46 loss 1.0995 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:39 loss 1.0771 (1.0673) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.023, TIME@all 0.304 +epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1131 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:39 loss 1.0534 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.987, TIME@all 0.304 +epoch: [320/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1228 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0518 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.154, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.0596 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.0589 (1.0609) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.849, TIME@all 0.304 +epoch: [321/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.1320 (1.0592) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.0705 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.904, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.1177 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.1212 (1.0639) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.906, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.1236 (1.0612) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0850 (1.0636) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.865, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.1159 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.1686 (1.0693) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.067, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.0807 (1.0623) acc 100.0000 (99.8438) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0941 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.017, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.0614 (1.0623) acc 100.0000 (99.6875) lr 0.000260 +epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.1331 (1.0703) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.891, TIME@all 0.304 +epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.0667 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0720 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.232, TIME@all 0.304 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0466 (1.0593) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0938 (1.0643) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.352, TIME@all 0.303 +epoch: [322/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0576 (1.0621) acc 100.0000 (99.6875) lr 0.000260 +epoch: [322/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0711 (1.0680) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.357, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0546 (1.0616) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.1261 (1.0736) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 844.370, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0932 (1.0596) acc 96.8750 (99.6875) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.1449 (1.0693) acc 96.8750 (99.7656) lr 0.000260 +FPS@all 844.320, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0709 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0823 (1.0664) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.319, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:07:14 loss 1.0678 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:07 loss 1.0947 (1.0688) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.508, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0520 (1.0669) acc 100.0000 (99.5312) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0922 (1.0760) acc 100.0000 (99.5312) lr 0.000260 +FPS@all 844.457, TIME@all 0.303 +epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0623 (1.0616) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 0:07:08 loss 1.0741 (1.0681) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.703, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:06:59 loss 1.0495 (1.0516) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:06:53 loss 1.0828 (1.0657) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.767, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0682 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0820 (1.0632) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.786, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:06:59 loss 1.0582 (1.0509) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:06:53 loss 1.0698 (1.0666) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.787, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0487 (1.0519) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0845 (1.0697) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.752, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:06:59 loss 1.0443 (1.0507) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:06:53 loss 1.1086 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.772, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0523 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0666 (1.0634) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.967, TIME@all 0.303 +epoch: [323/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0463 (1.0519) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0758 (1.0685) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.062, TIME@all 0.303 +epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0591 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0836 (1.0638) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.899, TIME@all 0.303 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0540 (1.0524) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0483 (1.0592) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.213, TIME@all 0.304 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0531 (1.0519) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0632 (1.0589) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.241, TIME@all 0.304 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0685 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0642 (1.0623) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.241, TIME@all 0.304 +epoch: [324/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0507 (1.0484) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0440 (1.0598) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.282, TIME@all 0.304 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0465 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.309 (0.304) data 0.001 (0.006) eta 0:06:38 loss 1.0483 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.330, TIME@all 0.304 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0929 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:38 loss 1.1219 (1.0636) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 843.437, TIME@all 0.304 +epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0612 (1.0668) acc 100.0000 (99.6875) lr 0.000260 +epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:38 loss 1.0863 (1.0670) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.391, TIME@all 0.304 +epoch: [324/350][20/50] time 0.312 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0554 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0792 (1.0607) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.556, TIME@all 0.303 +epoch: [325/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0690 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1474 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.539, TIME@all 0.303 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0489 (1.0617) acc 100.0000 (99.8438) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:23 loss 1.0740 (1.0671) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.542, TIME@all 0.303 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0643 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:23 loss 1.0659 (1.0610) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.458, TIME@all 0.304 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0631 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1672 (1.0666) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.504, TIME@all 0.303 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0882 (1.0617) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1385 (1.0694) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.489, TIME@all 0.304 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0677 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:06:23 loss 1.0898 (1.0669) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.840, TIME@all 0.303 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0642 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:06:23 loss 1.1432 (1.0681) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.680, TIME@all 0.303 +epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0521 (1.0578) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:23 loss 1.1264 (1.0691) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.634, TIME@all 0.303 +epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0530 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0607 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.555, TIME@all 0.304 +epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0613 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0945 (1.0612) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.632, TIME@all 0.304 +epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0518 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0510 (1.0614) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.590, TIME@all 0.304 +epoch: [326/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:06:14 loss 1.0540 (1.0519) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0854 (1.0623) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.543, TIME@all 0.304 +epoch: [326/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0475 (1.0575) acc 100.0000 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0582 (1.0664) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.749, TIME@all 0.304 +epoch: [326/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0503 (1.0581) acc 100.0000 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:06:08 loss 1.0564 (1.0708) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.551, TIME@all 0.304 +epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0778 (1.0552) acc 96.8750 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0461 (1.0621) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.071, TIME@all 0.304 +epoch: [326/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0547 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0562 (1.0615) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.879, TIME@all 0.304 +epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.1138 (1.0587) acc 100.0000 (99.8438) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0651 (1.0683) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.459, TIME@all 0.304 +epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0532 (1.0511) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0551 (1.0617) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.374, TIME@all 0.304 +epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0599 (1.0569) acc 100.0000 (99.8438) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.1292 (1.0676) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.420, TIME@all 0.304 +epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:05:58 loss 1.0559 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:53 loss 1.0560 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.607, TIME@all 0.304 +epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0527 (1.0534) acc 100.0000 (99.8438) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0606 (1.0614) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.415, TIME@all 0.304 +epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0576 (1.0552) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0929 (1.0654) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.412, TIME@all 0.304 +epoch: [327/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0679 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0577 (1.0695) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.796, TIME@all 0.304 +epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0693 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0508 (1.0630) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.545, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0722 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0696 (1.0678) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.089, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:05:44 loss 1.1244 (1.0637) acc 100.0000 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:37 loss 1.1079 (1.0711) acc 96.8750 (99.7656) lr 0.000260 +FPS@all 843.065, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.011) eta 0:05:44 loss 1.0770 (1.0592) acc 100.0000 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.1262 (1.0718) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.040, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:05:44 loss 1.1897 (1.0645) acc 96.8750 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:05:37 loss 1.0699 (1.0697) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.274, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0615 (1.0552) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0946 (1.0638) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.062, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0693 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.1054 (1.0704) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.066, TIME@all 0.304 +epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:05:44 loss 1.1414 (1.0617) acc 96.8750 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:37 loss 1.0959 (1.0677) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.210, TIME@all 0.304 +epoch: [328/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0559 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0846 (1.0767) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 843.380, TIME@all 0.304 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:05:27 loss 1.0837 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.006) eta 0:05:21 loss 1.0664 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.973, TIME@all 0.303 +epoch: [329/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:05:27 loss 1.0589 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.006) eta 0:05:21 loss 1.1538 (1.0637) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 844.919, TIME@all 0.303 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0547 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0489 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.932, TIME@all 0.303 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0990 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0696 (1.0682) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.938, TIME@all 0.303 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0717 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0645 (1.0630) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.933, TIME@all 0.303 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0705 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0504 (1.0618) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 845.093, TIME@all 0.303 +epoch: [329/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0817 (1.0714) acc 100.0000 (99.8438) lr 0.000260 +epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0654 (1.0720) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.333, TIME@all 0.303 +epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0558 (1.0500) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.291 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0495 (1.0604) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 845.154, TIME@all 0.303 +epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0555 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0550 (1.0643) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.964, TIME@all 0.304 +epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:05:12 loss 1.0755 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:07 loss 1.0459 (1.0623) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.027, TIME@all 0.304 +epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0623 (1.0565) acc 100.0000 (99.8438) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0449 (1.0649) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.960, TIME@all 0.304 +epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:05:12 loss 1.0473 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:07 loss 1.0452 (1.0731) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.936, TIME@all 0.304 +epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0474 (1.0549) acc 100.0000 (99.8438) lr 0.000260 +epoch: [330/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:06 loss 1.0792 (1.0635) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.146, TIME@all 0.304 +epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0702 (1.0571) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0625 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.111, TIME@all 0.304 +epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0610 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:05:07 loss 1.0522 (1.0659) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.945, TIME@all 0.304 +epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0505 (1.0593) acc 100.0000 (99.6875) lr 0.000260 +epoch: [330/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0500 (1.0647) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.146, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0656 (1.0618) acc 100.0000 (99.8438) lr 0.000260 +epoch: [331/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0512 (1.0697) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.456, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0623 (1.0559) acc 100.0000 (99.8438) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0550 (1.0651) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.557, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0714 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0574 (1.0661) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.693, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0511 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0575 (1.0674) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.464, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:58 loss 1.0566 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:04:52 loss 1.0733 (1.0634) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.658, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:58 loss 1.0960 (1.0601) acc 96.8750 (99.6875) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:04:52 loss 1.0553 (1.0695) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.666, TIME@all 0.304 +epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0781 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0553 (1.0702) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.472, TIME@all 0.304 +epoch: [331/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0529 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0804 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.841, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:04:43 loss 1.1200 (1.0605) acc 96.8750 (99.6875) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0581 (1.0710) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 842.270, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0904 (1.0599) acc 100.0000 (99.6875) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0456 (1.0727) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.322, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0498 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0429 (1.0731) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.381, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:04:43 loss 1.0475 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0560 (1.0657) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.285, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:04:43 loss 1.0534 (1.0566) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0774 (1.0699) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.504, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0786 (1.0593) acc 100.0000 (99.6875) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0613 (1.0709) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.306, TIME@all 0.304 +epoch: [332/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0623 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0461 (1.0674) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.702, TIME@all 0.304 +epoch: [332/350][20/50] time 0.307 (0.304) data 0.001 (0.014) eta 0:04:43 loss 1.0562 (1.0518) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0549 (1.0635) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.451, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:04:27 loss 1.0671 (1.0588) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:04:21 loss 1.0969 (1.0657) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.515, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0850 (1.0559) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0581 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.519, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:04:27 loss 1.0994 (1.0671) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:04:21 loss 1.0503 (1.0661) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.462, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0828 (1.0603) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0813 (1.0651) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.517, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:04:27 loss 1.0488 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0836 (1.0628) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.636, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0488 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0663 (1.0624) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.494, TIME@all 0.304 +epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:04:27 loss 1.0782 (1.0639) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0776 (1.0625) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.671, TIME@all 0.304 +epoch: [333/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0602 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0623 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.780, TIME@all 0.304 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0435 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0951 (1.0625) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.306, TIME@all 0.304 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0527 (1.0619) acc 100.0000 (99.6875) lr 0.000260 +epoch: [334/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0652 (1.0652) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.321, TIME@all 0.304 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0627 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0847 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.401, TIME@all 0.304 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0506 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0788 (1.0638) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.281, TIME@all 0.304 +epoch: [334/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0597 (1.0592) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0950 (1.0651) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.508, TIME@all 0.303 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.001 (0.014) eta 0:04:12 loss 1.0477 (1.0626) acc 100.0000 (99.8438) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0646 (1.0629) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.332, TIME@all 0.304 +epoch: [334/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0495 (1.0623) acc 100.0000 (99.8438) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0869 (1.0658) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.488, TIME@all 0.304 +epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0486 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.1123 (1.0607) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.698, TIME@all 0.303 +epoch: [335/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.1017 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0820 (1.0689) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.967, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0557 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0729 (1.0639) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.951, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 0:03:57 loss 1.0678 (1.0520) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.1025 (1.0600) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.895, TIME@all 0.304 +epoch: [335/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:03:57 loss 1.0545 (1.0570) acc 100.0000 (99.8438) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.0915 (1.0620) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.984, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0845 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0858 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.908, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0778 (1.0631) acc 100.0000 (99.8438) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:03:51 loss 1.1585 (1.0694) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.122, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:03:57 loss 1.0524 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.0800 (1.0646) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.234, TIME@all 0.304 +epoch: [335/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:03:57 loss 1.0891 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0613 (1.0650) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.086, TIME@all 0.304 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:03:41 loss 1.0726 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:03:35 loss 1.1752 (1.0650) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 844.144, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0582 (1.0536) acc 100.0000 (99.8438) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0749 (1.0646) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 844.230, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0458 (1.0510) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:03:35 loss 1.0525 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.244, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.1363 (1.0550) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0615 (1.0590) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.168, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0562 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0729 (1.0633) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.153, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:03:41 loss 1.1570 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.1174 (1.0632) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.358, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0753 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0756 (1.0640) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.308, TIME@all 0.303 +epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0620 (1.0523) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0577 (1.0599) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.475, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0655 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0800 (1.0697) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.842, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0521 (1.0584) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0667 (1.0662) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 845.872, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0880 (1.0686) acc 100.0000 (99.8438) lr 0.000260 +epoch: [337/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0803 (1.0733) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.909, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0700 (1.0632) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0611 (1.0663) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 845.863, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0619 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0526 (1.0699) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 845.880, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.001 (0.014) eta 0:03:26 loss 1.0519 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0905 (1.0727) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 846.058, TIME@all 0.303 +epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:03:26 loss 1.0801 (1.0633) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0950 (1.0715) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 846.025, TIME@all 0.303 +epoch: [337/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:03:26 loss 1.0560 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0774 (1.0696) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 846.078, TIME@all 0.303 +epoch: [338/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1471 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0615 (1.0668) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.566, TIME@all 0.303 +epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:03:11 loss 1.1256 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0554 (1.0601) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.622, TIME@all 0.303 +epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.0910 (1.0581) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0467 (1.0652) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.685, TIME@all 0.303 +epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1202 (1.0614) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:03:05 loss 1.0493 (1.0659) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.567, TIME@all 0.303 +epoch: [338/350][20/50] time 0.304 (0.304) data 0.001 (0.014) eta 0:03:11 loss 1.0806 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0446 (1.0595) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.799, TIME@all 0.303 +epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1176 (1.0629) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0609 (1.0682) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.600, TIME@all 0.303 +epoch: [338/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:03:11 loss 1.0640 (1.0543) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0789 (1.0608) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.043, TIME@all 0.303 +epoch: [338/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:03:11 loss 1.0684 (1.0578) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0611 (1.0665) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.726, TIME@all 0.303 +epoch: [339/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0415 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0533 (1.0655) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.093, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0466 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0508 (1.0658) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.122, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0605 (1.0499) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0520 (1.0584) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.141, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0562 (1.0534) acc 100.0000 (99.6875) lr 0.000260 +epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0544 (1.0624) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 842.090, TIME@all 0.304 +epoch: [339/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0544 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0814 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.114, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:02:56 loss 1.0742 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0495 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.427, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0556 (1.0583) acc 100.0000 (99.6875) lr 0.000260 +epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0475 (1.0627) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.310, TIME@all 0.304 +epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0576 (1.0566) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0454 (1.0654) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.269, TIME@all 0.304 +epoch: [340/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:02:40 loss 1.0626 (1.0534) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0470 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.942, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:02:40 loss 1.0780 (1.0592) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.1291 (1.0661) acc 96.8750 (99.7656) lr 0.000260 +FPS@all 844.115, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0517 (1.0537) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0437 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.971, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0894 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0595 (1.0630) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.989, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0751 (1.0578) acc 100.0000 (99.6875) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0616 (1.0688) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 843.951, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:02:40 loss 1.0668 (1.0599) acc 100.0000 (99.6875) lr 0.000260 +epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:34 loss 1.0517 (1.0605) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.147, TIME@all 0.303 +epoch: [340/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0612 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0568 (1.0570) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.397, TIME@all 0.303 +epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0892 (1.0586) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.1269 (1.0648) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 843.980, TIME@all 0.303 +epoch: [341/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:02:26 loss 1.0789 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:20 loss 1.1212 (1.0646) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.057, TIME@all 0.304 +epoch: [341/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0598 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:20 loss 1.0537 (1.0663) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.092, TIME@all 0.304 +epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.1190 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0751 (1.0633) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.126, TIME@all 0.304 +epoch: [341/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.1517 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0679 (1.0708) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.080, TIME@all 0.304 +epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0520 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:20 loss 1.0891 (1.0665) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.077, TIME@all 0.304 +epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:02:26 loss 1.0631 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0564 (1.0632) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.274, TIME@all 0.304 +epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0917 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0598 (1.0673) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 843.226, TIME@all 0.304 +epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0829 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.1178 (1.0708) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.474, TIME@all 0.304 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0532 (1.0567) acc 100.0000 (99.8438) lr 0.000260 +epoch: [342/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0885 (1.0630) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.753, TIME@all 0.303 +epoch: [342/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0464 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0717 (1.0656) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.789, TIME@all 0.303 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:02:10 loss 1.0506 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:02:04 loss 1.0597 (1.0651) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.829, TIME@all 0.303 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:02:10 loss 1.0534 (1.0542) acc 100.0000 (99.8438) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:02:04 loss 1.1029 (1.0685) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.744, TIME@all 0.303 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0877 (1.0562) acc 100.0000 (99.8438) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0855 (1.0670) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 843.752, TIME@all 0.303 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:02:10 loss 1.0494 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:04 loss 1.0859 (1.0630) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.920, TIME@all 0.303 +epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0537 (1.0610) acc 100.0000 (99.8438) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.1037 (1.0760) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 844.080, TIME@all 0.303 +epoch: [342/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:02:10 loss 1.0542 (1.0534) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:02:04 loss 1.0975 (1.0747) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 843.941, TIME@all 0.303 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.2145 (1.0600) acc 96.8750 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0720 (1.0634) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.588, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.1550 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0514 (1.0673) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.503, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.1553 (1.0548) acc 96.8750 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0487 (1.0656) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 842.642, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1276 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0684 (1.0682) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.527, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1111 (1.0588) acc 100.0000 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0706 (1.0639) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.672, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.0823 (1.0590) acc 100.0000 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0598 (1.0621) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.525, TIME@all 0.304 +epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:01:55 loss 1.1264 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0555 (1.0626) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.725, TIME@all 0.304 +epoch: [343/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1919 (1.0589) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0586 (1.0660) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.824, TIME@all 0.304 +epoch: [344/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0506 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0712 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.653, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:40 loss 1.0637 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:34 loss 1.0491 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.554, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0525 (1.0501) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0953 (1.0599) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.615, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0882 (1.0511) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0507 (1.0632) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.582, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0846 (1.0504) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0737 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.562, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0594 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0511 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.767, TIME@all 0.303 +epoch: [344/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0529 (1.0548) acc 100.0000 (99.8438) lr 0.000260 +epoch: [344/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0440 (1.0636) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.966, TIME@all 0.303 +epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0523 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0492 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.719, TIME@all 0.303 +epoch: [345/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.0863 (1.0556) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:18 loss 1.0439 (1.0625) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.171, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.0995 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:18 loss 1.0462 (1.0604) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.121, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.1178 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:18 loss 1.0489 (1.0650) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.194, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0824 (1.0504) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:01:18 loss 1.1278 (1.0634) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 844.137, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0948 (1.0522) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:01:18 loss 1.0696 (1.0621) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.336, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0892 (1.0630) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0532 (1.0677) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.143, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0968 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0832 (1.0620) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 844.284, TIME@all 0.303 +epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.2745 (1.0650) acc 96.8750 (99.8438) lr 0.000260 +epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0714 (1.0690) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 844.451, TIME@all 0.303 +epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0566 (1.0504) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1366 (1.0618) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.723, TIME@all 0.303 +epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0639 (1.0583) acc 100.0000 (99.8438) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1625 (1.0638) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 843.796, TIME@all 0.303 +epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0552 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:03 loss 1.0716 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.717, TIME@all 0.303 +epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:09 loss 1.0594 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.0958 (1.0577) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 843.793, TIME@all 0.303 +epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:09 loss 1.0578 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:03 loss 1.1800 (1.0616) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 843.944, TIME@all 0.303 +epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0536 (1.0549) acc 100.0000 (99.8438) lr 0.000260 +epoch: [346/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.0727 (1.0590) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 844.088, TIME@all 0.303 +epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0531 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1451 (1.0627) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.906, TIME@all 0.303 +epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0551 (1.0513) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:03 loss 1.0725 (1.0603) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.731, TIME@all 0.303 +epoch: [347/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0459 (1.0593) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0824 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.361, TIME@all 0.304 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0529 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0593 (1.0683) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.337, TIME@all 0.304 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0488 (1.0536) acc 100.0000 (99.8438) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0723 (1.0645) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.417, TIME@all 0.304 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0476 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0791 (1.0668) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.326, TIME@all 0.304 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0474 (1.0509) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0682 (1.0637) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.347, TIME@all 0.304 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:00:54 loss 1.0510 (1.0512) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0831 (1.0639) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.526, TIME@all 0.303 +epoch: [347/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0541 (1.0540) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:00:48 loss 1.0650 (1.0689) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.638, TIME@all 0.303 +epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0512 (1.0600) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0734 (1.0767) acc 100.0000 (99.5312) lr 0.000260 +FPS@all 843.505, TIME@all 0.303 +epoch: [348/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:00:39 loss 1.2009 (1.0585) acc 96.8750 (99.8438) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:00:33 loss 1.0688 (1.0711) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 842.848, TIME@all 0.304 +epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:00:39 loss 1.0993 (1.0502) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:00:33 loss 1.0607 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.955, TIME@all 0.304 +epoch: [348/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1287 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0657 (1.0636) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.883, TIME@all 0.304 +epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1586 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0507 (1.0664) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.887, TIME@all 0.304 +epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1240 (1.0571) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0503 (1.0677) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.878, TIME@all 0.304 +epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1111 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0784 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.029, TIME@all 0.304 +epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:00:39 loss 1.0875 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:00:33 loss 1.0564 (1.0620) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.101, TIME@all 0.304 +epoch: [348/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1682 (1.0571) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0693 (1.0653) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.205, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0618 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0585 (1.0622) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.800, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0487 (1.0562) acc 100.0000 (99.8438) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0740 (1.0619) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.917, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0497 (1.0596) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0446 (1.0663) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.756, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:00:24 loss 1.0556 (1.0591) acc 100.0000 (99.8438) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:00:18 loss 1.0548 (1.0634) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 842.739, TIME@all 0.304 +epoch: [349/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0463 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:00:18 loss 1.0492 (1.0653) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 842.818, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0612 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.1281 (1.0657) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 842.767, TIME@all 0.304 +epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0551 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0709 (1.0667) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 842.956, TIME@all 0.304 +epoch: [349/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0540 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0612 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.150, TIME@all 0.304 +epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:00:09 loss 1.0615 (1.0651) acc 100.0000 (99.6875) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0467 (1.0691) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 843.075, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:00:09 loss 1.0810 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0471 (1.0675) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.058, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0634 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0873 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 843.146, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.1015 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:00:03 loss 1.0493 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.089, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0629 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0624 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.078, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.1251 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0603 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 843.389, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0980 (1.0542) acc 100.0000 (99.8438) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0488 (1.0639) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.228, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:00:09 loss 1.0610 (1.0521) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0705 (1.0638) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 843.298, TIME@all 0.304 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 15913-by-512 matrix +Speed: 0.0301 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.3% +CMC curve +Rank-1 : 93.1% +Rank-5 : 97.6% +Rank-10 : 98.3% +Rank-20 : 99.0% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:44:48 +FPS@all 842.343, TIME@all 0.304 +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0302 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.2% +CMC curve +Rank-1 : 93.0% +Rank-5 : 97.5% +Rank-10 : 98.5% +Rank-20 : 99.0% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:44:53 +FPS@all 842.416, TIME@all 0.304 +Done, obtained 15913-by-512 matrix +Speed: 0.0301 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.3% +CMC curve +Rank-1 : 93.1% +Rank-5 : 97.6% +Rank-10 : 98.4% +Rank-20 : 99.2% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:44:55 +FPS@all 842.380, TIME@all 0.304 +Done, obtained 15913-by-512 matrix +Speed: 0.0302 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.2% +CMC curve +Rank-1 : 93.0% +Rank-5 : 97.4% +Rank-10 : 98.4% +Rank-20 : 99.1% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:44:57 +FPS@all 842.697, TIME@all 0.304 +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0312 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.2% +CMC curve +Rank-1 : 92.8% +Rank-5 : 97.5% +Rank-10 : 98.5% +Rank-20 : 99.0% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:45:00 +FPS@all 842.357, TIME@all 0.304 +Done, obtained 15913-by-512 matrix +Speed: 0.0310 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.2% +CMC curve +Rank-1 : 92.7% +Rank-5 : 97.6% +Rank-10 : 98.5% +Rank-20 : 99.1% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:45:00 +FPS@all 842.363, TIME@all 0.304 +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0319 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.2% +CMC curve +Rank-1 : 92.8% +Rank-5 : 97.5% +Rank-10 : 98.5% +Rank-20 : 99.1% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:45:01 +FPS@all 842.505, TIME@all 0.304 +Done, obtained 15913-by-512 matrix +Speed: 0.0333 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 80.3% +CMC curve +Rank-1 : 93.0% +Rank-5 : 97.5% +Rank-10 : 98.4% +Rank-20 : 99.1% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:45:04 +FPS@all 842.553, TIME@all 0.304 +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +THPModule_npu_shutdown success. +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 7 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 6 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 0 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 2 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 3 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 1 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 4 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Show configuration +adam: + beta1: 0.9 + beta2: 0.999 +addr: 127.0.0.1 +amp: True +cuhk03: + classic_split: False + labeled_images: False + use_metric_cuhk03: False +data: + combineall: False + height: 256 + k_tfm: 1 + load_train_targets: False + norm_mean: [0.485, 0.456, 0.406] + norm_std: [0.229, 0.224, 0.225] + root: reid-data + save_dir: log/osnet_x1_0_market1501_softmax + sources: ['market1501'] + split_id: 0 + targets: ['market1501'] + transforms: ['random_flip', 'random_crop', 'random_patch'] + type: image + width: 128 + workers: 4 +device_num: 8 +ignore_classifer: False +local_rank: 5 +loss: + name: softmax + softmax: + label_smooth: True + triplet: + margin: 0.3 + weight_t: 1.0 + weight_x: 0.0 +market1501: + use_500k_distractors: False +model: + load_weights: + name: osnet_x1_0 + pretrained: False + resume: +rmsprop: + alpha: 0.99 +sampler: + num_cams: 1 + num_datasets: 1 + num_instances: 4 + train_sampler: RandomSampler + train_sampler_t: RandomSampler +sgd: + dampening: 0.0 + momentum: 0.9 + nesterov: False +test: + batch_size: 300 + dist_metric: euclidean + eval_freq: -1 + evaluate: False + normalize_feature: False + ranks: [1, 5, 10, 20] + rerank: False + start_eval: 300 + visrank: False + visrank_topk: 10 +train: + base_lr_mult: 0.1 + batch_size: 32 + fixbase_epoch: 0 + gamma: 0.1 + lr: 0.26 + lr_scheduler: multi_step + max_epoch: 350 + new_layers: ['classifier'] + open_layers: ['classifier'] + optim: sgd + print_freq: 20 + seed: 1 + staged_lr: False + start_epoch: 0 + stepsize: [150, 225, 300] + weight_decay: 0.0005 +use_gpu: False +use_npu: True +video: + pooling_method: avg + sample_method: evenly + seq_len: 15 + +Collecting env info ... +** System info ** +PyTorch version: 1.8.1+ascend.rc2 +Is debug build: False +CUDA used to build PyTorch: None +ROCM used to build PyTorch: N/A + +OS: CentOS Linux 7 (AltArch) (aarch64) +GCC version: (GCC) 7.3.0 +Clang version: 3.9.1 (tags/RELEASE_391/final) +CMake version: version 3.18.6 + +Python version: 3.7 (64-bit runtime) +Is CUDA available: False +CUDA runtime version: No CUDA +GPU models and configuration: No CUDA +Nvidia driver version: No CUDA +cuDNN version: No CUDA +HIP runtime version: N/A +MIOpen runtime version: N/A + +Versions of relevant libraries: +[pip3] numpy==1.20.0 +[pip3] torch==1.8.1+ascend.rc2.20220505 +[pip3] torch-npu==1.8.1rc2.post20220505 +[pip3] torchreid==1.4.0 +[conda] numpy 1.20.0 +[conda] torch 1.8.1+ascend.rc2.20220505 +[conda] torch-npu 1.8.1rc2.post20220505 +[conda] torchreid 1.4.0 + Pillow (8.4.0) + +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. + +Defaults for this optimization level are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : dynamic +combine_grad : None +combine_ddp : None +ddp_replica_count : 4 +check_combined_tensors : None +user_cast_preferred : None +Processing user overrides (additional kwargs that are not None)... +After processing overrides, optimization options are: +enabled : True +opt_level : O2 +cast_model_type : torch.float16 +patch_torch_functions : False +keep_batchnorm_fp32 : True +master_weights : True +loss_scale : dynamic +combine_grad : True +combine_ddp : None +ddp_replica_count : 4 +check_combined_tensors : None +user_cast_preferred : None +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +Building train transforms ... ++ resize to 256x128 ++ random flip ++ random crop (enlarge to 288x144 and crop 256x128) ++ random patch ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +Building test transforms ... ++ resize to 256x128 ++ to torch tensor of range [0, 1] ++ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) +=> Loading train (source) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- +=> Loading test (target) dataset +=> Loaded Market1501 + ---------------------------------------- + subset | # ids | # images | # cameras + ---------------------------------------- + train | 751 | 12936 | 6 + query | 750 | 3368 | 6 + gallery | 751 | 15913 | 6 + ---------------------------------------- + + + **************** Summary **************** + source : ['market1501'] + # source datasets : 1 + # source ids : 751 + # source images : 12936 + # source cameras : 6 + target : ['market1501'] + ***************************************** + + +Building model: osnet_x1_0 +Model complexity: params=2,193,616 flops=978,878,352 +Use npu fused optimizer +Building softmax-engine for image-reid +=> Start training +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. +grad.sizes() = [751, 512], strides() = [512, 1] +bucket_view.sizes() = [385024], strides() = [1] (function operator()) +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.316 (0.311) data 0.000 (0.020) eta 1:30:42 loss 6.6282 (6.6135) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/350][40/50] time 0.315 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.0348 (6.4896) acc 6.2500 (0.8594) lr 0.260000 +FPS@all 821.678, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.8040 (6.5953) acc 0.0000 (0.4688) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1929 (6.4627) acc 0.0000 (0.9375) lr 0.260000 +FPS@all 821.595, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:43 loss 6.6438 (6.6055) acc 0.0000 (0.3125) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1450 (6.4875) acc 6.2500 (0.5469) lr 0.260000 +FPS@all 821.637, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.019) eta 1:30:43 loss 6.2564 (6.6608) acc 3.1250 (0.6250) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.2591 (6.5006) acc 3.1250 (0.9375) lr 0.260000 +FPS@all 821.562, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.4853 (6.6085) acc 0.0000 (0.4688) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.1861 (6.4820) acc 0.0000 (0.8594) lr 0.260000 +FPS@all 821.564, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.7360 (6.6453) acc 0.0000 (0.6250) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1298 (6.4847) acc 0.0000 (0.9375) lr 0.260000 +FPS@all 821.594, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.5267 (6.6343) acc 0.0000 (0.7812) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.001 (0.010) eta 1:30:44 loss 5.9071 (6.4698) acc 6.2500 (1.0156) lr 0.260000 +FPS@all 821.580, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +group num: 1 +epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:43 loss 6.4037 (6.6935) acc 0.0000 (0.4688) lr 0.260000 +epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.2472 (6.5596) acc 0.0000 (0.7031) lr 0.260000 +FPS@all 821.601, TIME@all 0.312 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 6.0594 (5.7936) acc 3.1250 (3.2812) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.7203 (5.7094) acc 6.2500 (3.9844) lr 0.260000 +FPS@all 819.110, TIME@all 0.313 +epoch: [2/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 1:31:09 loss 5.9182 (5.6758) acc 0.0000 (3.7500) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:53 loss 5.6438 (5.6485) acc 9.3750 (4.6875) lr 0.260000 +FPS@all 819.211, TIME@all 0.312 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.5824 (5.7873) acc 12.5000 (3.1250) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.3965 (5.6825) acc 3.1250 (4.3750) lr 0.260000 +FPS@all 819.078, TIME@all 0.313 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.5807 (5.6932) acc 0.0000 (4.3750) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.5829 (5.6599) acc 6.2500 (3.6719) lr 0.260000 +FPS@all 819.054, TIME@all 0.313 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:31:10 loss 6.0790 (5.7108) acc 0.0000 (4.0625) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:54 loss 5.3312 (5.6364) acc 6.2500 (4.2969) lr 0.260000 +FPS@all 819.147, TIME@all 0.313 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:31:10 loss 5.8013 (5.6971) acc 3.1250 (4.6875) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:30:54 loss 5.3562 (5.6617) acc 9.3750 (4.5312) lr 0.260000 +FPS@all 819.131, TIME@all 0.313 +epoch: [2/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 1:31:12 loss 5.9211 (5.7517) acc 3.1250 (4.0625) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:54 loss 5.2705 (5.6378) acc 9.3750 (4.8438) lr 0.260000 +FPS@all 819.092, TIME@all 0.313 +epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.7639 (5.7057) acc 3.1250 (4.3750) lr 0.260000 +epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 6.0995 (5.6769) acc 3.1250 (4.6875) lr 0.260000 +FPS@all 819.107, TIME@all 0.313 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.2832 (5.0634) acc 12.5000 (9.8438) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.9239 (5.0338) acc 12.5000 (10.3906) lr 0.260000 +FPS@all 821.691, TIME@all 0.312 +epoch: [3/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:30:26 loss 5.0421 (5.0929) acc 9.3750 (8.1250) lr 0.260000 +epoch: [3/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:30:18 loss 5.1250 (5.1040) acc 6.2500 (8.8281) lr 0.260000 +FPS@all 821.794, TIME@all 0.312 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3481 (5.0394) acc 9.3750 (8.5938) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 5.3156 (5.0389) acc 6.2500 (9.6875) lr 0.260000 +FPS@all 821.637, TIME@all 0.312 +epoch: [3/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:30:28 loss 5.3440 (5.0542) acc 6.2500 (7.8125) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 5.1863 (5.0536) acc 6.2500 (8.5938) lr 0.260000 +FPS@all 821.704, TIME@all 0.312 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3779 (5.0746) acc 3.1250 (8.5938) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.5517 (5.0395) acc 18.7500 (10.0000) lr 0.260000 +FPS@all 821.575, TIME@all 0.312 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3214 (5.0822) acc 3.1250 (7.1875) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.8877 (5.0807) acc 15.6250 (7.5000) lr 0.260000 +FPS@all 821.692, TIME@all 0.312 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.1941 (5.0012) acc 6.2500 (6.5625) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.5551 (4.9956) acc 12.5000 (8.9844) lr 0.260000 +FPS@all 821.684, TIME@all 0.312 +epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:30:28 loss 5.2215 (5.0933) acc 9.3750 (9.5312) lr 0.260000 +epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:30:19 loss 4.7463 (5.0586) acc 12.5000 (9.9219) lr 0.260000 +FPS@all 821.633, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:30:08 loss 4.5543 (4.3096) acc 18.7500 (19.6875) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:03 loss 4.8111 (4.4200) acc 15.6250 (18.2812) lr 0.260000 +FPS@all 821.298, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:30:08 loss 4.6653 (4.3899) acc 15.6250 (17.8125) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.4343 (4.4955) acc 21.8750 (17.6562) lr 0.260000 +FPS@all 821.246, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.2725 (4.3375) acc 25.0000 (18.7500) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.7870 (4.4286) acc 18.7500 (17.8125) lr 0.260000 +FPS@all 821.185, TIME@all 0.312 +epoch: [4/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5120 (4.3763) acc 9.3750 (17.1875) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.5298 (4.4602) acc 18.7500 (18.0469) lr 0.260000 +FPS@all 821.231, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.6118 (4.3939) acc 15.6250 (18.1250) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:30:04 loss 4.8010 (4.4485) acc 18.7500 (17.9688) lr 0.260000 +FPS@all 821.188, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5778 (4.4145) acc 21.8750 (17.9688) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:30:04 loss 4.2522 (4.4577) acc 25.0000 (17.4219) lr 0.260000 +FPS@all 821.161, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.1779 (4.4533) acc 21.8750 (17.8125) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.7549 (4.4869) acc 9.3750 (17.1094) lr 0.260000 +FPS@all 821.213, TIME@all 0.312 +epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5193 (4.3964) acc 9.3750 (15.7812) lr 0.260000 +epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.0433 (4.4299) acc 31.2500 (17.6562) lr 0.260000 +FPS@all 821.152, TIME@all 0.312 +epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:53 loss 3.8829 (3.7926) acc 31.2500 (30.0000) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.0074 (3.8962) acc 25.0000 (29.3750) lr 0.260000 +FPS@all 822.299, TIME@all 0.311 +epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:52 loss 3.4718 (3.8024) acc 43.7500 (30.0000) lr 0.260000 +epoch: [5/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:29:40 loss 3.6481 (3.8641) acc 37.5000 (28.9844) lr 0.260000 +FPS@all 822.400, TIME@all 0.311 +epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:29:55 loss 4.1549 (3.7249) acc 18.7500 (31.0938) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:29:41 loss 4.3441 (3.8644) acc 15.6250 (28.3594) lr 0.260000 +FPS@all 822.233, TIME@all 0.311 +epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:54 loss 3.8879 (3.7821) acc 25.0000 (30.1562) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.6259 (3.8353) acc 43.7500 (31.2500) lr 0.260000 +FPS@all 822.243, TIME@all 0.311 +epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:54 loss 4.4040 (3.7438) acc 12.5000 (26.8750) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.7219 (3.8507) acc 28.1250 (28.0469) lr 0.260000 +FPS@all 822.306, TIME@all 0.311 +epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:53 loss 4.3460 (3.8087) acc 25.0000 (27.8125) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.2156 (3.8984) acc 37.5000 (28.7500) lr 0.260000 +FPS@all 822.309, TIME@all 0.311 +epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:55 loss 3.5324 (3.8007) acc 37.5000 (30.9375) lr 0.260000 +epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.9986 (3.8975) acc 18.7500 (29.2188) lr 0.260000 +FPS@all 822.305, TIME@all 0.311 +epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:54 loss 4.3092 (3.8331) acc 15.6250 (29.8438) lr 0.260000 +epoch: [5/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.0275 (3.8858) acc 31.2500 (29.7656) lr 0.260000 +FPS@all 822.267, TIME@all 0.311 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:29:30 loss 3.9603 (3.4243) acc 28.1250 (40.7812) lr 0.260000 +epoch: [6/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:29:31 loss 3.7473 (3.4712) acc 34.3750 (40.0000) lr 0.260000 +FPS@all 821.414, TIME@all 0.312 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:29:29 loss 3.9790 (3.3634) acc 18.7500 (38.4375) lr 0.260000 +epoch: [6/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:29:30 loss 3.8342 (3.4920) acc 25.0000 (36.9531) lr 0.260000 +FPS@all 821.463, TIME@all 0.312 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7156 (3.2911) acc 34.3750 (40.9375) lr 0.260000 +epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.3102 (3.4258) acc 40.6250 (39.6875) lr 0.260000 +FPS@all 821.298, TIME@all 0.312 +epoch: [6/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:29:30 loss 3.5310 (3.3215) acc 46.8750 (41.8750) lr 0.260000 +epoch: [6/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:29:31 loss 3.5656 (3.4785) acc 34.3750 (37.8906) lr 0.260000 +FPS@all 821.316, TIME@all 0.312 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7731 (3.3322) acc 40.6250 (40.3125) lr 0.260000 +epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.5444 (3.4172) acc 40.6250 (40.1562) lr 0.260000 +FPS@all 821.248, TIME@all 0.312 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7150 (3.3124) acc 31.2500 (42.6562) lr 0.260000 +epoch: [6/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.4702 (3.4333) acc 34.3750 (38.9062) lr 0.260000 +FPS@all 821.333, TIME@all 0.312 +epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.6914 (3.3798) acc 34.3750 (38.7500) lr 0.260000 +epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.8998 (3.5084) acc 34.3750 (36.0938) lr 0.260000 +FPS@all 821.341, TIME@all 0.312 +epoch: [6/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:29:30 loss 3.4229 (3.2996) acc 43.7500 (43.1250) lr 0.260000 +epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:29:31 loss 3.7600 (3.4180) acc 40.6250 (40.2344) lr 0.260000 +FPS@all 821.329, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.014) eta 1:29:40 loss 2.6181 (2.8738) acc 59.3750 (55.6250) lr 0.260000 +epoch: [7/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:29:24 loss 3.1618 (3.0069) acc 50.0000 (51.4844) lr 0.260000 +FPS@all 820.658, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.014) eta 1:29:40 loss 2.6422 (2.8396) acc 59.3750 (52.6562) lr 0.260000 +epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.2460 (2.9835) acc 50.0000 (50.7031) lr 0.260000 +FPS@all 820.598, TIME@all 0.312 +epoch: [7/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:29:41 loss 2.9682 (2.9077) acc 56.2500 (53.5938) lr 0.260000 +epoch: [7/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.8760 (3.0461) acc 28.1250 (50.5469) lr 0.260000 +FPS@all 820.447, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:29:40 loss 3.0837 (2.8961) acc 50.0000 (53.1250) lr 0.260000 +epoch: [7/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.3221 (3.0226) acc 43.7500 (51.1719) lr 0.260000 +FPS@all 820.540, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 1:29:41 loss 3.0217 (2.8466) acc 50.0000 (53.4375) lr 0.260000 +epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:29:25 loss 3.4676 (2.9847) acc 43.7500 (51.1719) lr 0.260000 +FPS@all 820.478, TIME@all 0.312 +epoch: [7/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:29:41 loss 3.2219 (2.8566) acc 43.7500 (53.5938) lr 0.260000 +epoch: [7/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.3742 (2.9948) acc 53.1250 (52.0312) lr 0.260000 +FPS@all 820.515, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:29:40 loss 3.0862 (2.8851) acc 28.1250 (50.0000) lr 0.260000 +epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.4040 (3.0138) acc 46.8750 (48.9062) lr 0.260000 +FPS@all 820.538, TIME@all 0.312 +epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 1:29:41 loss 2.7115 (2.8474) acc 43.7500 (52.8125) lr 0.260000 +epoch: [7/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.2168 (3.0193) acc 40.6250 (49.5312) lr 0.260000 +FPS@all 820.532, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:58 loss 2.8560 (2.6150) acc 59.3750 (63.1250) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.8085 (2.7530) acc 56.2500 (58.2812) lr 0.260000 +FPS@all 821.782, TIME@all 0.312 +epoch: [8/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:00 loss 3.1258 (2.6251) acc 50.0000 (59.5312) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.9904 (2.7642) acc 50.0000 (56.2500) lr 0.260000 +FPS@all 821.743, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:28:59 loss 3.2261 (2.5946) acc 43.7500 (61.8750) lr 0.260000 +epoch: [8/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:28:59 loss 2.8836 (2.7469) acc 50.0000 (58.2031) lr 0.260000 +FPS@all 821.580, TIME@all 0.312 +epoch: [8/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:28:59 loss 3.1327 (2.5782) acc 53.1250 (62.0312) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:59 loss 2.7518 (2.7192) acc 59.3750 (58.6719) lr 0.260000 +FPS@all 821.669, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:28:59 loss 3.1808 (2.6771) acc 43.7500 (57.6562) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:28:58 loss 2.8134 (2.8056) acc 50.0000 (55.0781) lr 0.260000 +FPS@all 821.619, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:28:57 loss 3.0446 (2.5873) acc 53.1250 (59.5312) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:28:57 loss 3.0669 (2.7482) acc 46.8750 (56.3281) lr 0.260000 +FPS@all 821.778, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:59 loss 2.9624 (2.5745) acc 40.6250 (60.9375) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.9526 (2.7848) acc 59.3750 (56.4844) lr 0.260000 +FPS@all 821.656, TIME@all 0.312 +epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:59 loss 3.0653 (2.6125) acc 65.6250 (61.2500) lr 0.260000 +epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.6907 (2.7866) acc 59.3750 (55.9375) lr 0.260000 +FPS@all 821.671, TIME@all 0.312 +epoch: [9/350][20/50] time 0.313 (0.311) data 0.000 (0.013) eta 1:28:33 loss 3.2726 (2.4585) acc 53.1250 (66.2500) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.0010 (2.6279) acc 50.0000 (62.9688) lr 0.260000 +FPS@all 822.341, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.2206 (2.3918) acc 53.1250 (68.4375) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.1333 (2.5713) acc 50.0000 (63.9062) lr 0.260000 +FPS@all 822.369, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.0911 (2.4511) acc 46.8750 (65.6250) lr 0.260000 +epoch: [9/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8350 (2.5984) acc 56.2500 (62.3438) lr 0.260000 +FPS@all 822.282, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 2.9704 (2.4047) acc 53.1250 (67.8125) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8338 (2.5180) acc 56.2500 (64.1406) lr 0.260000 +FPS@all 822.243, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.1379 (2.4459) acc 53.1250 (66.8750) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8686 (2.5407) acc 59.3750 (63.8281) lr 0.260000 +FPS@all 822.252, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 2.8515 (2.4064) acc 65.6250 (68.5938) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.9643 (2.5307) acc 53.1250 (64.2188) lr 0.260000 +FPS@all 822.313, TIME@all 0.311 +epoch: [9/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.0585 (2.5614) acc 37.5000 (60.4688) lr 0.260000 +epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.7029 (2.6364) acc 65.6250 (60.2344) lr 0.260000 +FPS@all 822.290, TIME@all 0.311 +epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.3398 (2.4733) acc 46.8750 (65.0000) lr 0.260000 +epoch: [9/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.2641 (2.5778) acc 34.3750 (63.1250) lr 0.260000 +FPS@all 822.314, TIME@all 0.311 +epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:43 loss 2.3517 (2.2324) acc 68.7500 (70.6250) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:28:31 loss 2.4602 (2.3565) acc 75.0000 (68.9062) lr 0.260000 +FPS@all 821.370, TIME@all 0.312 +epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.6665 (2.2341) acc 71.8750 (71.4062) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4884 (2.3236) acc 75.0000 (69.7656) lr 0.260000 +FPS@all 821.271, TIME@all 0.312 +epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.5749 (2.1301) acc 62.5000 (74.6875) lr 0.260000 +epoch: [10/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:28:33 loss 2.2472 (2.2847) acc 78.1250 (71.6406) lr 0.260000 +FPS@all 821.225, TIME@all 0.312 +epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:28:43 loss 2.4498 (2.1611) acc 71.8750 (73.2812) lr 0.260000 +epoch: [10/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:28:32 loss 2.0609 (2.2643) acc 78.1250 (71.2500) lr 0.260000 +FPS@all 821.249, TIME@all 0.312 +epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.6028 (2.2636) acc 59.3750 (70.0000) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4973 (2.3759) acc 71.8750 (68.2812) lr 0.260000 +FPS@all 821.222, TIME@all 0.312 +epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.5421 (2.1689) acc 65.6250 (71.4062) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.3444 (2.3500) acc 65.6250 (66.3281) lr 0.260000 +FPS@all 821.268, TIME@all 0.312 +epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.3957 (2.1793) acc 59.3750 (72.1875) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4514 (2.3398) acc 78.1250 (68.9844) lr 0.260000 +FPS@all 821.277, TIME@all 0.312 +epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.2044 (2.1960) acc 68.7500 (72.8125) lr 0.260000 +epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.5738 (2.3373) acc 59.3750 (68.5938) lr 0.260000 +FPS@all 821.241, TIME@all 0.312 +epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:26 loss 2.3666 (2.0413) acc 68.7500 (77.6562) lr 0.260000 +epoch: [11/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0087 (2.1680) acc 75.0000 (75.0000) lr 0.260000 +FPS@all 821.105, TIME@all 0.312 +epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:26 loss 2.1604 (2.0000) acc 71.8750 (80.0000) lr 0.260000 +epoch: [11/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0525 (2.1788) acc 84.3750 (75.0781) lr 0.260000 +FPS@all 821.184, TIME@all 0.312 +epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.011) eta 1:28:27 loss 2.4809 (2.0056) acc 78.1250 (80.0000) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0819 (2.1647) acc 75.0000 (75.0781) lr 0.260000 +FPS@all 821.026, TIME@all 0.312 +epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:28 loss 1.8513 (1.9640) acc 84.3750 (80.9375) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:18 loss 2.3437 (2.1515) acc 71.8750 (77.0312) lr 0.260000 +FPS@all 820.962, TIME@all 0.312 +epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.5034 (1.9836) acc 62.5000 (80.6250) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2583 (2.1649) acc 75.0000 (75.3125) lr 0.260000 +FPS@all 821.041, TIME@all 0.312 +epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.4249 (2.0324) acc 65.6250 (77.3438) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.1741 (2.1452) acc 71.8750 (75.2344) lr 0.260000 +FPS@all 821.095, TIME@all 0.312 +epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 1.9815 (1.9537) acc 78.1250 (80.4688) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2618 (2.1193) acc 71.8750 (76.0156) lr 0.260000 +FPS@all 821.061, TIME@all 0.312 +epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.2706 (2.0041) acc 65.6250 (78.4375) lr 0.260000 +epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2561 (2.1261) acc 75.0000 (76.2500) lr 0.260000 +FPS@all 821.005, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:11 loss 2.3362 (1.9903) acc 71.8750 (79.6875) lr 0.260000 +epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:28:00 loss 2.3193 (2.1579) acc 59.3750 (75.1562) lr 0.260000 +FPS@all 821.060, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:10 loss 2.6404 (1.9881) acc 71.8750 (79.2188) lr 0.260000 +epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:28:00 loss 2.2368 (2.1016) acc 75.0000 (76.1719) lr 0.260000 +FPS@all 821.081, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:10 loss 2.1621 (1.8695) acc 81.2500 (83.2812) lr 0.260000 +epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.3304 (2.0960) acc 68.7500 (76.3281) lr 0.260000 +FPS@all 820.956, TIME@all 0.312 +epoch: [12/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:28:10 loss 2.2149 (2.0055) acc 81.2500 (79.8438) lr 0.260000 +epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.3125 (2.0941) acc 68.7500 (76.2500) lr 0.260000 +FPS@all 820.968, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:28:10 loss 2.4423 (1.9677) acc 71.8750 (79.3750) lr 0.260000 +epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.2852 (2.1149) acc 71.8750 (75.5469) lr 0.260000 +FPS@all 821.024, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:10 loss 2.4935 (1.9875) acc 71.8750 (80.3125) lr 0.260000 +epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.5237 (2.1232) acc 62.5000 (76.3281) lr 0.260000 +FPS@all 820.987, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:10 loss 2.2782 (1.9396) acc 71.8750 (81.4062) lr 0.260000 +epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.0881 (2.1168) acc 78.1250 (76.1719) lr 0.260000 +FPS@all 820.981, TIME@all 0.312 +epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:10 loss 2.1213 (1.9307) acc 78.1250 (82.5000) lr 0.260000 +epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:28:01 loss 2.1073 (2.0569) acc 71.8750 (78.0469) lr 0.260000 +FPS@all 820.952, TIME@all 0.312 +epoch: [13/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:27:41 loss 1.9837 (1.9019) acc 78.1250 (83.7500) lr 0.260000 +epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:27:44 loss 1.9490 (1.9490) acc 84.3750 (81.2500) lr 0.260000 +FPS@all 821.521, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:27:42 loss 2.1825 (1.9398) acc 71.8750 (81.7188) lr 0.260000 +epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:27:44 loss 2.0111 (2.0049) acc 78.1250 (78.5156) lr 0.260000 +FPS@all 821.514, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:44 loss 2.4584 (1.9165) acc 71.8750 (82.6562) lr 0.260000 +epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.8755 (1.9853) acc 87.5000 (80.1562) lr 0.260000 +FPS@all 821.324, TIME@all 0.312 +epoch: [13/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:27:42 loss 2.0462 (1.9351) acc 78.1250 (80.3125) lr 0.260000 +epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:27:44 loss 2.2325 (1.9893) acc 71.8750 (78.9844) lr 0.260000 +FPS@all 821.473, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:42 loss 2.0369 (1.9011) acc 75.0000 (82.5000) lr 0.260000 +epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.8497 (1.9845) acc 81.2500 (80.3906) lr 0.260000 +FPS@all 821.452, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:43 loss 2.3142 (1.9082) acc 65.6250 (83.5938) lr 0.260000 +epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.9372 (1.9584) acc 78.1250 (81.3281) lr 0.260000 +FPS@all 821.433, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:43 loss 2.3263 (1.9224) acc 71.8750 (83.7500) lr 0.260000 +epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 2.0749 (1.9442) acc 71.8750 (81.2500) lr 0.260000 +FPS@all 821.408, TIME@all 0.312 +epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:27:42 loss 2.2744 (1.9465) acc 75.0000 (80.6250) lr 0.260000 +epoch: [13/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 1:27:45 loss 2.0181 (1.9905) acc 71.8750 (79.0625) lr 0.260000 +FPS@all 821.437, TIME@all 0.312 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.1549 (1.7656) acc 78.1250 (86.0938) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 2.0388 (1.8583) acc 75.0000 (84.4531) lr 0.260000 +FPS@all 818.548, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:47 loss 2.1645 (1.7621) acc 65.6250 (86.0938) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:34 loss 1.8989 (1.8531) acc 87.5000 (83.4375) lr 0.260000 +FPS@all 818.604, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.0675 (1.7825) acc 84.3750 (84.8438) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:36 loss 2.3007 (1.8602) acc 78.1250 (82.9688) lr 0.260000 +FPS@all 818.478, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 1:27:49 loss 2.0988 (1.8254) acc 81.2500 (84.2188) lr 0.260000 +epoch: [14/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 1:27:36 loss 2.0766 (1.8895) acc 75.0000 (81.7969) lr 0.260000 +FPS@all 818.491, TIME@all 0.313 +epoch: [14/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2654 (1.8558) acc 71.8750 (82.6562) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 1.9649 (1.9041) acc 78.1250 (81.0156) lr 0.260000 +FPS@all 818.487, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2270 (1.8433) acc 75.0000 (82.9688) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.001 (0.006) eta 1:27:35 loss 2.2333 (1.9130) acc 68.7500 (80.9375) lr 0.260000 +FPS@all 818.520, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2180 (1.7951) acc 71.8750 (85.6250) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 2.3446 (1.8823) acc 71.8750 (83.0469) lr 0.260000 +FPS@all 818.488, TIME@all 0.313 +epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:48 loss 1.9552 (1.7729) acc 78.1250 (85.9375) lr 0.260000 +epoch: [14/350][40/50] time 0.316 (0.313) data 0.001 (0.006) eta 1:27:35 loss 2.1505 (1.8777) acc 78.1250 (82.5000) lr 0.260000 +FPS@all 818.507, TIME@all 0.313 +epoch: [15/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 1:27:32 loss 1.8626 (1.7616) acc 78.1250 (86.8750) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.4890 (1.8177) acc 96.8750 (84.8438) lr 0.260000 +FPS@all 819.971, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:32 loss 1.5863 (1.7754) acc 93.7500 (88.2812) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.7300 (1.8268) acc 84.3750 (84.6875) lr 0.260000 +FPS@all 820.012, TIME@all 0.312 +epoch: [15/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 1:27:33 loss 2.0121 (1.7987) acc 81.2500 (85.6250) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.6598 (1.8514) acc 90.6250 (83.3594) lr 0.260000 +FPS@all 819.902, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.001 (0.012) eta 1:27:33 loss 1.8165 (1.7817) acc 78.1250 (86.2500) lr 0.260000 +epoch: [15/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:27:18 loss 1.8319 (1.8493) acc 84.3750 (83.9844) lr 0.260000 +FPS@all 819.937, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.011) eta 1:27:33 loss 2.0262 (1.7596) acc 81.2500 (86.2500) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.8683 (1.8221) acc 81.2500 (84.7656) lr 0.260000 +FPS@all 819.876, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:33 loss 1.9889 (1.8129) acc 75.0000 (83.5938) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.7582 (1.8574) acc 87.5000 (82.7344) lr 0.260000 +FPS@all 819.899, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:33 loss 1.6255 (1.7841) acc 84.3750 (85.3125) lr 0.260000 +epoch: [15/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:27:18 loss 2.2408 (1.8380) acc 75.0000 (83.4375) lr 0.260000 +FPS@all 819.946, TIME@all 0.312 +epoch: [15/350][20/50] time 0.318 (0.313) data 0.001 (0.012) eta 1:27:33 loss 1.9661 (1.7990) acc 81.2500 (86.0938) lr 0.260000 +epoch: [15/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:27:18 loss 1.8645 (1.8606) acc 90.6250 (83.5938) lr 0.260000 +FPS@all 819.898, TIME@all 0.312 +epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.9626 (1.7379) acc 81.2500 (86.7188) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 2.0224 (1.7724) acc 78.1250 (85.4688) lr 0.260000 +FPS@all 821.089, TIME@all 0.312 +epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:26:58 loss 1.7966 (1.6555) acc 81.2500 (87.9688) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.9531 (1.7421) acc 71.8750 (85.7812) lr 0.260000 +FPS@all 821.115, TIME@all 0.312 +epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6974 (1.6741) acc 84.3750 (88.1250) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.8325 (1.7259) acc 84.3750 (86.7188) lr 0.260000 +FPS@all 821.033, TIME@all 0.312 +epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6967 (1.7068) acc 87.5000 (86.7188) lr 0.260000 +epoch: [16/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.7892 (1.7877) acc 87.5000 (84.7656) lr 0.260000 +FPS@all 821.024, TIME@all 0.312 +epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 2.0385 (1.6987) acc 78.1250 (89.0625) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 2.0935 (1.8025) acc 75.0000 (84.2188) lr 0.260000 +FPS@all 821.073, TIME@all 0.312 +epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 2.0115 (1.6608) acc 87.5000 (89.8438) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.8627 (1.7365) acc 87.5000 (87.3438) lr 0.260000 +FPS@all 821.063, TIME@all 0.312 +epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6344 (1.6227) acc 90.6250 (90.4688) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.8369 (1.7179) acc 81.2500 (87.1094) lr 0.260000 +FPS@all 821.047, TIME@all 0.312 +epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.9252 (1.6842) acc 84.3750 (88.7500) lr 0.260000 +epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.9735 (1.7510) acc 75.0000 (86.9531) lr 0.260000 +FPS@all 821.036, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:26:55 loss 1.7763 (1.6445) acc 84.3750 (89.8438) lr 0.260000 +epoch: [17/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:26:47 loss 1.8996 (1.6952) acc 78.1250 (87.5781) lr 0.260000 +FPS@all 820.376, TIME@all 0.312 +epoch: [17/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:26:56 loss 1.8223 (1.6305) acc 75.0000 (89.2188) lr 0.260000 +epoch: [17/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:26:48 loss 1.8678 (1.6984) acc 81.2500 (87.1875) lr 0.260000 +FPS@all 820.305, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.7350 (1.6289) acc 81.2500 (89.5312) lr 0.260000 +epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.8668 (1.6902) acc 84.3750 (87.6562) lr 0.260000 +FPS@all 820.253, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.8903 (1.6255) acc 84.3750 (89.6875) lr 0.260000 +epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.8793 (1.6891) acc 78.1250 (88.2031) lr 0.260000 +FPS@all 820.244, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:57 loss 1.7764 (1.6301) acc 87.5000 (90.6250) lr 0.260000 +epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.9442 (1.6842) acc 75.0000 (88.3594) lr 0.260000 +FPS@all 820.269, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:57 loss 1.6899 (1.6538) acc 87.5000 (89.5312) lr 0.260000 +epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.5833 (1.7040) acc 96.8750 (88.0469) lr 0.260000 +FPS@all 820.281, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:26:57 loss 1.6075 (1.6221) acc 87.5000 (89.5312) lr 0.260000 +epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.7865 (1.6844) acc 84.3750 (87.9688) lr 0.260000 +FPS@all 820.276, TIME@all 0.312 +epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.4329 (1.6028) acc 96.8750 (91.8750) lr 0.260000 +epoch: [17/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.7390 (1.6717) acc 87.5000 (89.7656) lr 0.260000 +FPS@all 820.282, TIME@all 0.312 +epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:24 loss 1.6715 (1.5873) acc 90.6250 (90.6250) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6752 (1.6425) acc 84.3750 (88.6719) lr 0.260000 +FPS@all 821.144, TIME@all 0.312 +epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.5557 (1.6220) acc 90.6250 (89.6875) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6507 (1.6201) acc 90.6250 (90.7031) lr 0.260000 +FPS@all 821.050, TIME@all 0.312 +epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.6027 (1.5640) acc 90.6250 (91.2500) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:24 loss 1.5662 (1.6056) acc 90.6250 (90.0781) lr 0.260000 +FPS@all 820.961, TIME@all 0.312 +epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.6935 (1.5632) acc 90.6250 (92.6562) lr 0.260000 +epoch: [18/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6172 (1.6251) acc 93.7500 (90.7031) lr 0.260000 +FPS@all 821.026, TIME@all 0.312 +epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:26:24 loss 1.8983 (1.5618) acc 81.2500 (91.7188) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.5856 (1.5760) acc 93.7500 (90.8594) lr 0.260000 +FPS@all 821.064, TIME@all 0.312 +epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.5269 (1.5832) acc 90.6250 (91.5625) lr 0.260000 +epoch: [18/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:26:23 loss 1.5916 (1.5909) acc 90.6250 (91.0156) lr 0.260000 +FPS@all 821.000, TIME@all 0.312 +epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:26:24 loss 1.6521 (1.5482) acc 90.6250 (91.8750) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.001 (0.007) eta 1:26:23 loss 1.4857 (1.5833) acc 93.7500 (90.4688) lr 0.260000 +FPS@all 821.040, TIME@all 0.312 +epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:24 loss 1.7063 (1.5715) acc 90.6250 (90.9375) lr 0.260000 +epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.5518 (1.5956) acc 87.5000 (90.0000) lr 0.260000 +FPS@all 821.079, TIME@all 0.312 +epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:09 loss 2.3830 (1.6386) acc 65.6250 (88.7500) lr 0.260000 +epoch: [19/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:26:07 loss 1.5716 (1.6989) acc 87.5000 (86.9531) lr 0.260000 +FPS@all 821.271, TIME@all 0.312 +epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.8563 (1.5330) acc 84.3750 (92.9688) lr 0.260000 +epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.4016 (1.5983) acc 100.0000 (91.0156) lr 0.260000 +FPS@all 821.255, TIME@all 0.312 +epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:09 loss 1.5695 (1.6285) acc 93.7500 (90.1562) lr 0.260000 +epoch: [19/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.5778 (1.6661) acc 90.6250 (88.3594) lr 0.260000 +FPS@all 821.136, TIME@all 0.312 +epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.5845 (1.5691) acc 93.7500 (91.5625) lr 0.260000 +epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.6185 (1.6599) acc 90.6250 (88.5938) lr 0.260000 +FPS@all 821.141, TIME@all 0.312 +epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.9166 (1.6418) acc 81.2500 (89.5312) lr 0.260000 +epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.4058 (1.6638) acc 100.0000 (88.3594) lr 0.260000 +FPS@all 821.146, TIME@all 0.312 +epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:26:10 loss 1.6911 (1.5734) acc 87.5000 (91.5625) lr 0.260000 +epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:26:09 loss 1.6064 (1.6249) acc 90.6250 (89.8438) lr 0.260000 +FPS@all 821.071, TIME@all 0.312 +epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.7532 (1.5856) acc 81.2500 (90.6250) lr 0.260000 +epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.8280 (1.6303) acc 93.7500 (90.0000) lr 0.260000 +FPS@all 821.117, TIME@all 0.312 +epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.8908 (1.5553) acc 81.2500 (92.6562) lr 0.260000 +epoch: [19/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:26:09 loss 1.6892 (1.6409) acc 87.5000 (89.5312) lr 0.260000 +FPS@all 821.165, TIME@all 0.312 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:25:41 loss 1.7847 (1.5536) acc 84.3750 (90.3125) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.007) eta 1:25:41 loss 1.7815 (1.6190) acc 87.5000 (88.7500) lr 0.260000 +FPS@all 823.080, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.6803 (1.5365) acc 87.5000 (91.2500) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.8391 (1.5954) acc 87.5000 (90.2344) lr 0.260000 +FPS@all 823.147, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:42 loss 1.6136 (1.5561) acc 84.3750 (91.4062) lr 0.260000 +epoch: [20/350][40/50] time 0.309 (0.311) data 0.000 (0.006) eta 1:25:42 loss 1.5522 (1.6205) acc 90.6250 (90.0781) lr 0.260000 +FPS@all 822.994, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:42 loss 1.6710 (1.5545) acc 93.7500 (92.5000) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:42 loss 1.6654 (1.6119) acc 87.5000 (90.0781) lr 0.260000 +FPS@all 823.016, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.7364 (1.6071) acc 87.5000 (89.8438) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.5424 (1.6320) acc 87.5000 (89.2188) lr 0.260000 +FPS@all 823.022, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.6891 (1.5813) acc 87.5000 (90.7812) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.7291 (1.6304) acc 84.3750 (89.3750) lr 0.260000 +FPS@all 823.037, TIME@all 0.311 +epoch: [20/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:25:43 loss 1.6846 (1.5600) acc 84.3750 (92.0312) lr 0.260000 +epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.6395 (1.6013) acc 90.6250 (90.8594) lr 0.260000 +FPS@all 823.048, TIME@all 0.311 +epoch: [20/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:25:41 loss 1.5817 (1.5890) acc 90.6250 (91.4062) lr 0.260000 +epoch: [20/350][40/50] time 0.309 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.4465 (1.6364) acc 90.6250 (89.1406) lr 0.260000 +FPS@all 823.023, TIME@all 0.311 +epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.7868 (1.5410) acc 87.5000 (91.8750) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:51 loss 1.5713 (1.5955) acc 93.7500 (90.6250) lr 0.260000 +FPS@all 819.334, TIME@all 0.312 +epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.7525 (1.5567) acc 81.2500 (91.7188) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:52 loss 1.6842 (1.5877) acc 81.2500 (90.4688) lr 0.260000 +FPS@all 819.247, TIME@all 0.312 +epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.5501 (1.5733) acc 90.6250 (89.3750) lr 0.260000 +epoch: [21/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.8139 (1.6260) acc 81.2500 (88.1250) lr 0.260000 +FPS@all 819.206, TIME@all 0.312 +epoch: [21/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:25:50 loss 1.5850 (1.5124) acc 93.7500 (93.4375) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.7471 (1.5828) acc 84.3750 (91.7188) lr 0.260000 +FPS@all 819.188, TIME@all 0.313 +epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.5787 (1.5192) acc 90.6250 (93.7500) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:52 loss 1.5839 (1.5715) acc 90.6250 (91.9531) lr 0.260000 +FPS@all 819.215, TIME@all 0.312 +epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.6615 (1.5549) acc 90.6250 (92.9688) lr 0.260000 +epoch: [21/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.5989 (1.6092) acc 93.7500 (91.0938) lr 0.260000 +FPS@all 819.231, TIME@all 0.312 +epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.7835 (1.5895) acc 84.3750 (91.0938) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:25:52 loss 1.8190 (1.6289) acc 81.2500 (89.3750) lr 0.260000 +FPS@all 819.220, TIME@all 0.312 +epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.4475 (1.5246) acc 96.8750 (93.1250) lr 0.260000 +epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:52 loss 1.5312 (1.5960) acc 96.8750 (90.4688) lr 0.260000 +FPS@all 819.230, TIME@all 0.312 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:26 loss 1.5565 (1.5155) acc 93.7500 (92.5000) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.6679 (1.5861) acc 90.6250 (90.4688) lr 0.260000 +FPS@all 822.416, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:26 loss 1.7279 (1.4898) acc 87.5000 (94.2188) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:13 loss 1.6047 (1.5567) acc 90.6250 (91.7188) lr 0.260000 +FPS@all 822.483, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:28 loss 1.6697 (1.5038) acc 75.0000 (92.9688) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:25:14 loss 1.6614 (1.5683) acc 87.5000 (91.8750) lr 0.260000 +FPS@all 822.309, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.6048 (1.4859) acc 93.7500 (93.7500) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.4808 (1.5697) acc 93.7500 (91.4062) lr 0.260000 +FPS@all 822.360, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.5731 (1.5341) acc 90.6250 (93.5938) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:25:14 loss 1.6963 (1.5518) acc 84.3750 (92.1094) lr 0.260000 +FPS@all 822.343, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:27 loss 1.5886 (1.5144) acc 93.7500 (92.6562) lr 0.260000 +epoch: [22/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.6312 (1.5789) acc 87.5000 (91.0938) lr 0.260000 +FPS@all 822.345, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.5611 (1.5027) acc 84.3750 (92.5000) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.7899 (1.5576) acc 81.2500 (91.3281) lr 0.260000 +FPS@all 822.277, TIME@all 0.311 +epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.4891 (1.5125) acc 96.8750 (93.7500) lr 0.260000 +epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.5369 (1.5716) acc 96.8750 (91.7188) lr 0.260000 +FPS@all 822.345, TIME@all 0.311 +epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:25:17 loss 1.7702 (1.5403) acc 87.5000 (91.2500) lr 0.260000 +epoch: [23/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:21 loss 1.5015 (1.5722) acc 90.6250 (90.6250) lr 0.260000 +FPS@all 819.361, TIME@all 0.312 +epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:25:18 loss 1.6664 (1.5323) acc 87.5000 (92.6562) lr 0.260000 +epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4780 (1.5620) acc 90.6250 (91.4062) lr 0.260000 +FPS@all 819.254, TIME@all 0.312 +epoch: [23/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:25:18 loss 1.5097 (1.5438) acc 90.6250 (90.9375) lr 0.260000 +epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:22 loss 1.4420 (1.5536) acc 90.6250 (91.0938) lr 0.260000 +FPS@all 819.160, TIME@all 0.313 +epoch: [23/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:25:19 loss 1.7214 (1.5280) acc 87.5000 (91.7188) lr 0.260000 +epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:22 loss 1.6131 (1.5598) acc 90.6250 (90.5469) lr 0.260000 +FPS@all 819.181, TIME@all 0.313 +epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:18 loss 1.4078 (1.5490) acc 96.8750 (91.4062) lr 0.260000 +epoch: [23/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4210 (1.5597) acc 96.8750 (91.7969) lr 0.260000 +FPS@all 819.304, TIME@all 0.312 +epoch: [23/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:25:18 loss 1.6521 (1.5040) acc 90.6250 (93.1250) lr 0.260000 +epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4713 (1.5296) acc 96.8750 (91.7188) lr 0.260000 +FPS@all 819.207, TIME@all 0.312 +epoch: [23/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:25:18 loss 1.4584 (1.5363) acc 93.7500 (92.6562) lr 0.260000 +epoch: [23/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:25:22 loss 1.4382 (1.5483) acc 93.7500 (92.1875) lr 0.260000 +FPS@all 819.208, TIME@all 0.312 +epoch: [23/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:25:18 loss 1.6535 (1.4719) acc 84.3750 (94.3750) lr 0.260000 +epoch: [23/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:22 loss 1.6647 (1.5356) acc 84.3750 (92.5000) lr 0.260000 +FPS@all 819.261, TIME@all 0.312 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4750 (1.4499) acc 93.7500 (94.8438) lr 0.260000 +epoch: [24/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:08 loss 1.7962 (1.5007) acc 75.0000 (92.9688) lr 0.260000 +FPS@all 818.933, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.5692 (1.4956) acc 87.5000 (92.8125) lr 0.260000 +epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.5767 (1.4912) acc 90.6250 (93.2812) lr 0.260000 +FPS@all 818.874, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:25:06 loss 1.3935 (1.4988) acc 93.7500 (94.5312) lr 0.260000 +epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:09 loss 1.6542 (1.5023) acc 87.5000 (93.8281) lr 0.260000 +FPS@all 818.844, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:07 loss 1.5128 (1.4872) acc 93.7500 (93.4375) lr 0.260000 +epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.4863 (1.5115) acc 93.7500 (93.1250) lr 0.260000 +FPS@all 818.728, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4324 (1.4779) acc 93.7500 (94.0625) lr 0.260000 +epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.3412 (1.5241) acc 96.8750 (92.8906) lr 0.260000 +FPS@all 818.815, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:25:07 loss 1.3886 (1.5098) acc 96.8750 (92.9688) lr 0.260000 +epoch: [24/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:25:09 loss 1.6375 (1.5465) acc 93.7500 (92.6562) lr 0.260000 +FPS@all 818.733, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4676 (1.4700) acc 93.7500 (95.3125) lr 0.260000 +epoch: [24/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.8077 (1.5185) acc 84.3750 (92.7344) lr 0.260000 +FPS@all 818.784, TIME@all 0.313 +epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.5346 (1.4835) acc 93.7500 (94.2188) lr 0.260000 +epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.5902 (1.5221) acc 90.6250 (92.5781) lr 0.260000 +FPS@all 818.801, TIME@all 0.313 +epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:24:36 loss 1.6575 (1.5398) acc 93.7500 (91.2500) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:24:35 loss 1.4669 (1.5589) acc 90.6250 (90.0781) lr 0.260000 +FPS@all 821.650, TIME@all 0.312 +epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:24:38 loss 1.6323 (1.5345) acc 90.6250 (93.7500) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.5432 (1.5545) acc 90.6250 (91.9531) lr 0.260000 +FPS@all 821.533, TIME@all 0.312 +epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:24:37 loss 1.5416 (1.4944) acc 90.6250 (92.6562) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:24:36 loss 1.3653 (1.5414) acc 100.0000 (91.3281) lr 0.260000 +FPS@all 821.570, TIME@all 0.312 +epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:24:38 loss 1.5862 (1.5524) acc 93.7500 (90.1562) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4607 (1.5549) acc 93.7500 (91.4844) lr 0.260000 +FPS@all 821.489, TIME@all 0.312 +epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.5154 (1.4999) acc 90.6250 (93.2812) lr 0.260000 +epoch: [25/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4267 (1.5069) acc 90.6250 (93.2031) lr 0.260000 +FPS@all 821.524, TIME@all 0.312 +epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.5492 (1.4890) acc 87.5000 (93.7500) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4962 (1.5225) acc 93.7500 (92.5781) lr 0.260000 +FPS@all 821.550, TIME@all 0.312 +epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.4934 (1.4866) acc 93.7500 (93.9062) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4729 (1.5239) acc 87.5000 (92.4219) lr 0.260000 +FPS@all 821.565, TIME@all 0.312 +epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:36 loss 1.6209 (1.5181) acc 87.5000 (92.3438) lr 0.260000 +epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:35 loss 1.5020 (1.5536) acc 93.7500 (91.4062) lr 0.260000 +FPS@all 821.589, TIME@all 0.312 +epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.5639 (1.4421) acc 90.6250 (94.0625) lr 0.260000 +epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.4490 (1.5070) acc 96.8750 (92.1094) lr 0.260000 +FPS@all 819.451, TIME@all 0.312 +epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:40 loss 1.4258 (1.4228) acc 96.8750 (95.7812) lr 0.260000 +epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:31 loss 1.4979 (1.4618) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 819.304, TIME@all 0.312 +epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:38 loss 1.4228 (1.4274) acc 96.8750 (94.3750) lr 0.260000 +epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.5030 (1.4775) acc 90.6250 (93.3594) lr 0.260000 +FPS@all 819.418, TIME@all 0.312 +epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:24:39 loss 1.5766 (1.4439) acc 90.6250 (93.7500) lr 0.260000 +epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:24:31 loss 1.4421 (1.4967) acc 96.8750 (92.9688) lr 0.260000 +FPS@all 819.304, TIME@all 0.312 +epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.4220 (1.4260) acc 100.0000 (95.4688) lr 0.260000 +epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.5720 (1.4863) acc 90.6250 (92.9688) lr 0.260000 +FPS@all 819.365, TIME@all 0.312 +epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.5034 (1.4309) acc 96.8750 (94.8438) lr 0.260000 +epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.3989 (1.4764) acc 96.8750 (93.6719) lr 0.260000 +FPS@all 819.371, TIME@all 0.312 +epoch: [26/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:24:39 loss 1.5285 (1.4272) acc 93.7500 (95.9375) lr 0.260000 +epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:24:30 loss 1.5201 (1.4971) acc 96.8750 (93.6719) lr 0.260000 +FPS@all 819.381, TIME@all 0.312 +epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:40 loss 1.5282 (1.4895) acc 87.5000 (93.1250) lr 0.260000 +epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:31 loss 1.6047 (1.5262) acc 90.6250 (92.5000) lr 0.260000 +FPS@all 819.318, TIME@all 0.312 +epoch: [27/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 1:24:14 loss 1.3199 (1.4374) acc 100.0000 (94.3750) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:00 loss 1.6500 (1.4866) acc 87.5000 (93.2812) lr 0.260000 +FPS@all 821.657, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 1:24:15 loss 1.4118 (1.4190) acc 96.8750 (95.3125) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:24:01 loss 1.5913 (1.4713) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 821.489, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5150 (1.4186) acc 90.6250 (95.1562) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.5893 (1.4833) acc 84.3750 (93.4375) lr 0.260000 +FPS@all 821.460, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:14 loss 1.5717 (1.4449) acc 90.6250 (95.0000) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:00 loss 1.5136 (1.5098) acc 93.7500 (93.0469) lr 0.260000 +FPS@all 821.544, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5690 (1.4168) acc 93.7500 (95.6250) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.6238 (1.4746) acc 93.7500 (93.9844) lr 0.260000 +FPS@all 821.480, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.4096 (1.4111) acc 96.8750 (95.4688) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.5977 (1.4766) acc 90.6250 (93.4375) lr 0.260000 +FPS@all 821.510, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5878 (1.4378) acc 87.5000 (94.2188) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.4629 (1.4715) acc 93.7500 (93.5938) lr 0.260000 +FPS@all 821.515, TIME@all 0.312 +epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5008 (1.4471) acc 90.6250 (94.2188) lr 0.260000 +epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.6759 (1.4845) acc 84.3750 (93.7500) lr 0.260000 +FPS@all 821.505, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:34 loss 1.3526 (1.4557) acc 96.8750 (94.8438) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:43 loss 1.3564 (1.4703) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 821.865, TIME@all 0.311 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:34 loss 1.3073 (1.4758) acc 100.0000 (94.5312) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:43 loss 1.5193 (1.5056) acc 90.6250 (93.0469) lr 0.260000 +FPS@all 821.779, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:35 loss 1.3095 (1.4245) acc 100.0000 (95.7812) lr 0.260000 +epoch: [28/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.3887 (1.4792) acc 90.6250 (93.6719) lr 0.260000 +FPS@all 821.713, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.011) eta 1:23:35 loss 1.4862 (1.4779) acc 93.7500 (94.2188) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.3751 (1.5002) acc 96.8750 (93.0469) lr 0.260000 +FPS@all 821.697, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:23:35 loss 1.6803 (1.4677) acc 87.5000 (93.7500) lr 0.260000 +epoch: [28/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5154 (1.4982) acc 93.7500 (93.0469) lr 0.260000 +FPS@all 821.727, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:35 loss 1.5152 (1.4850) acc 90.6250 (93.2812) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5318 (1.5160) acc 93.7500 (92.4219) lr 0.260000 +FPS@all 821.748, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:23:35 loss 1.3699 (1.4909) acc 96.8750 (93.7500) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5084 (1.5226) acc 93.7500 (91.8750) lr 0.260000 +FPS@all 821.711, TIME@all 0.312 +epoch: [28/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:23:35 loss 1.5735 (1.4754) acc 87.5000 (94.0625) lr 0.260000 +epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.4265 (1.5083) acc 100.0000 (93.6719) lr 0.260000 +FPS@all 821.757, TIME@all 0.312 +epoch: [29/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.4936 (1.4160) acc 90.6250 (95.0000) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3840 (1.4551) acc 100.0000 (94.1406) lr 0.260000 +FPS@all 820.661, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.4282 (1.4216) acc 90.6250 (95.3125) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:36 loss 1.5305 (1.4687) acc 90.6250 (92.7344) lr 0.260000 +FPS@all 820.671, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.3515 (1.4073) acc 96.8750 (94.8438) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3879 (1.4298) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 820.557, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:23:46 loss 1.3265 (1.4087) acc 100.0000 (95.0000) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:23:37 loss 1.3521 (1.4509) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 820.540, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.2630 (1.3978) acc 96.8750 (95.4688) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.4642 (1.4492) acc 96.8750 (94.0625) lr 0.260000 +FPS@all 820.561, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.4438 (1.4284) acc 93.7500 (93.7500) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3917 (1.4376) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 820.556, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.4322 (1.4140) acc 96.8750 (95.0000) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.5395 (1.4669) acc 90.6250 (93.8281) lr 0.260000 +FPS@all 820.612, TIME@all 0.312 +epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.5072 (1.4242) acc 93.7500 (94.8438) lr 0.260000 +epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3888 (1.4453) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 820.637, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:23:40 loss 1.3873 (1.4740) acc 93.7500 (93.4375) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.5961 (1.5236) acc 87.5000 (91.7188) lr 0.260000 +FPS@all 820.589, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.4927 (1.4637) acc 96.8750 (93.1250) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.6540 (1.4897) acc 84.3750 (92.8125) lr 0.260000 +FPS@all 820.555, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:40 loss 1.6383 (1.4638) acc 90.6250 (94.0625) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:25 loss 1.6774 (1.5285) acc 84.3750 (92.1875) lr 0.260000 +FPS@all 820.485, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:39 loss 1.5117 (1.4890) acc 93.7500 (92.9688) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:25 loss 1.4164 (1.5279) acc 93.7500 (91.6406) lr 0.260000 +FPS@all 820.457, TIME@all 0.312 +epoch: [30/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.5782 (1.4623) acc 87.5000 (95.1562) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:25 loss 1.5256 (1.5118) acc 93.7500 (92.5000) lr 0.260000 +FPS@all 820.497, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:40 loss 1.5389 (1.4561) acc 90.6250 (94.0625) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:24 loss 1.4736 (1.5317) acc 93.7500 (91.7188) lr 0.260000 +FPS@all 820.491, TIME@all 0.312 +epoch: [30/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.4243 (1.4673) acc 93.7500 (93.2812) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.4600 (1.5348) acc 90.6250 (91.3281) lr 0.260000 +FPS@all 820.511, TIME@all 0.312 +epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:39 loss 1.4943 (1.4753) acc 96.8750 (94.0625) lr 0.260000 +epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:24 loss 1.5571 (1.5056) acc 87.5000 (92.8125) lr 0.260000 +FPS@all 820.451, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4845 (1.4211) acc 93.7500 (93.9062) lr 0.260000 +epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.5522 (1.4661) acc 84.3750 (92.5781) lr 0.260000 +FPS@all 820.527, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4855 (1.4177) acc 96.8750 (95.1562) lr 0.260000 +epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:05 loss 1.4288 (1.4582) acc 100.0000 (93.6719) lr 0.260000 +FPS@all 820.590, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:23:08 loss 1.5682 (1.4101) acc 87.5000 (96.0938) lr 0.260000 +epoch: [31/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 1:23:06 loss 1.4159 (1.4545) acc 100.0000 (94.3750) lr 0.260000 +FPS@all 820.489, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 1:23:08 loss 1.4143 (1.4311) acc 96.8750 (95.0000) lr 0.260000 +epoch: [31/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.2883 (1.4577) acc 100.0000 (94.4531) lr 0.260000 +FPS@all 820.498, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.6776 (1.4463) acc 81.2500 (93.9062) lr 0.260000 +epoch: [31/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.4019 (1.4725) acc 100.0000 (92.7344) lr 0.260000 +FPS@all 820.488, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.3520 (1.4116) acc 96.8750 (96.0938) lr 0.260000 +epoch: [31/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:23:06 loss 1.3501 (1.4434) acc 96.8750 (95.0781) lr 0.260000 +FPS@all 820.498, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4087 (1.4098) acc 96.8750 (95.6250) lr 0.260000 +epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:07 loss 1.4141 (1.4671) acc 96.8750 (94.1406) lr 0.260000 +FPS@all 820.436, TIME@all 0.312 +epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4699 (1.4447) acc 93.7500 (95.4688) lr 0.260000 +epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.3289 (1.4611) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 820.486, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:22:57 loss 1.6850 (1.3778) acc 84.3750 (95.3125) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:22:43 loss 1.4879 (1.4139) acc 96.8750 (94.6094) lr 0.260000 +FPS@all 821.433, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.8145 (1.4029) acc 84.3750 (95.1562) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3301 (1.4341) acc 100.0000 (94.4531) lr 0.260000 +FPS@all 821.460, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.6815 (1.3823) acc 93.7500 (96.7188) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.3002 (1.4056) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 821.337, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:22:58 loss 1.5301 (1.3848) acc 90.6250 (96.8750) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.4035 (1.4140) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 821.307, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.4965 (1.3613) acc 90.6250 (97.3438) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.4128 (1.4080) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 821.358, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.5097 (1.4131) acc 87.5000 (94.6875) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3919 (1.4353) acc 93.7500 (93.9062) lr 0.260000 +FPS@all 821.358, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.3657 (1.3704) acc 93.7500 (95.3125) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3849 (1.4004) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 821.366, TIME@all 0.312 +epoch: [32/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:22:58 loss 1.6577 (1.3664) acc 93.7500 (96.8750) lr 0.260000 +epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.4665 (1.4010) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 821.328, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:22:43 loss 1.3181 (1.4284) acc 100.0000 (95.6250) lr 0.260000 +epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.4651 (1.4819) acc 90.6250 (93.5156) lr 0.260000 +FPS@all 820.371, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:22:42 loss 1.3704 (1.4187) acc 96.8750 (95.4688) lr 0.260000 +epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.4664 (1.4792) acc 90.6250 (93.0469) lr 0.260000 +FPS@all 820.446, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:43 loss 1.6915 (1.4567) acc 90.6250 (95.0000) lr 0.260000 +epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.3825 (1.4818) acc 100.0000 (93.7500) lr 0.260000 +FPS@all 820.281, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:44 loss 1.3712 (1.4253) acc 93.7500 (95.0000) lr 0.260000 +epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:22:42 loss 1.4577 (1.4523) acc 90.6250 (94.2969) lr 0.260000 +FPS@all 820.244, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:42 loss 1.3153 (1.4308) acc 100.0000 (95.1562) lr 0.260000 +epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.3471 (1.4584) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 820.379, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:44 loss 1.4781 (1.4555) acc 93.7500 (94.0625) lr 0.260000 +epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.5819 (1.4809) acc 90.6250 (93.4375) lr 0.260000 +FPS@all 820.269, TIME@all 0.312 +epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:43 loss 1.5082 (1.4165) acc 96.8750 (95.7812) lr 0.260000 +epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.4395 (1.4579) acc 96.8750 (94.4531) lr 0.260000 +FPS@all 820.283, TIME@all 0.312 +epoch: [33/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:22:44 loss 1.4600 (1.4722) acc 93.7500 (93.7500) lr 0.260000 +epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.5006 (1.4922) acc 87.5000 (93.3594) lr 0.260000 +FPS@all 820.307, TIME@all 0.312 +epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3004 (1.4298) acc 96.8750 (95.0000) lr 0.260000 +epoch: [34/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:22:23 loss 1.8499 (1.4629) acc 87.5000 (93.4375) lr 0.260000 +FPS@all 820.182, TIME@all 0.312 +epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:27 loss 1.3592 (1.4146) acc 96.8750 (95.4688) lr 0.260000 +epoch: [34/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:22:23 loss 1.5750 (1.4391) acc 93.7500 (95.0000) lr 0.260000 +FPS@all 820.241, TIME@all 0.312 +epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3915 (1.4168) acc 100.0000 (95.3125) lr 0.260000 +epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.5279 (1.4492) acc 87.5000 (94.6094) lr 0.260000 +FPS@all 820.073, TIME@all 0.312 +epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.2267 (1.4572) acc 100.0000 (94.2188) lr 0.260000 +epoch: [34/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:22:24 loss 1.3501 (1.4694) acc 93.7500 (93.9844) lr 0.260000 +FPS@all 820.104, TIME@all 0.312 +epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:27 loss 1.3010 (1.4026) acc 96.8750 (96.0938) lr 0.260000 +epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.6941 (1.4517) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 820.155, TIME@all 0.312 +epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3584 (1.4042) acc 96.8750 (95.3125) lr 0.260000 +epoch: [34/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:22:24 loss 1.4948 (1.4070) acc 90.6250 (95.1562) lr 0.260000 +FPS@all 820.117, TIME@all 0.312 +epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.4699 (1.4144) acc 93.7500 (95.4688) lr 0.260000 +epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.5347 (1.4448) acc 90.6250 (94.4531) lr 0.260000 +FPS@all 820.114, TIME@all 0.312 +epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.4314 (1.4114) acc 90.6250 (95.1562) lr 0.260000 +epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.4645 (1.4184) acc 90.6250 (94.9219) lr 0.260000 +FPS@all 820.151, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.3476 (1.4101) acc 100.0000 (95.0000) lr 0.260000 +epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:53 loss 1.3865 (1.4248) acc 96.8750 (94.6094) lr 0.260000 +FPS@all 821.718, TIME@all 0.312 +epoch: [35/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.3780 (1.4115) acc 96.8750 (95.1562) lr 0.260000 +epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:21:53 loss 1.3565 (1.4177) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 821.792, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.001 (0.013) eta 1:21:51 loss 1.4937 (1.3989) acc 90.6250 (96.4062) lr 0.260000 +epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.7045 (1.4260) acc 84.3750 (94.8438) lr 0.260000 +FPS@all 821.675, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.5457 (1.4135) acc 93.7500 (95.4688) lr 0.260000 +epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.4825 (1.4364) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 821.702, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:21:51 loss 1.3611 (1.4208) acc 96.8750 (94.3750) lr 0.260000 +epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.4827 (1.4307) acc 96.8750 (94.3750) lr 0.260000 +FPS@all 821.653, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:21:51 loss 1.4476 (1.4162) acc 96.8750 (96.0938) lr 0.260000 +epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.6511 (1.4354) acc 87.5000 (94.9219) lr 0.260000 +FPS@all 821.694, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.5204 (1.4219) acc 93.7500 (94.6875) lr 0.260000 +epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.4591 (1.4398) acc 90.6250 (94.1406) lr 0.260000 +FPS@all 821.686, TIME@all 0.312 +epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:52 loss 1.5611 (1.4349) acc 90.6250 (93.5938) lr 0.260000 +epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.5331 (1.4318) acc 90.6250 (93.9844) lr 0.260000 +FPS@all 821.688, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3962 (1.3926) acc 93.7500 (95.3125) lr 0.260000 +epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3841 (1.4200) acc 93.7500 (94.2188) lr 0.260000 +FPS@all 820.421, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3782 (1.3898) acc 93.7500 (95.3125) lr 0.260000 +epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.5343 (1.4287) acc 93.7500 (94.3750) lr 0.260000 +FPS@all 820.443, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.4774 (1.3566) acc 90.6250 (96.5625) lr 0.260000 +epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:52 loss 1.2916 (1.4113) acc 96.8750 (94.8438) lr 0.260000 +FPS@all 820.289, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:21:54 loss 1.4740 (1.3851) acc 96.8750 (95.3125) lr 0.260000 +epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.4540 (1.4247) acc 96.8750 (94.2188) lr 0.260000 +FPS@all 820.332, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.5673 (1.4176) acc 90.6250 (95.1562) lr 0.260000 +epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3714 (1.4209) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 820.365, TIME@all 0.312 +epoch: [36/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:21:55 loss 1.5204 (1.4149) acc 93.7500 (95.1562) lr 0.260000 +epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:21:51 loss 1.3479 (1.4381) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 820.298, TIME@all 0.312 +epoch: [36/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:21:55 loss 1.3733 (1.3760) acc 96.8750 (95.7812) lr 0.260000 +epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3131 (1.4046) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 820.331, TIME@all 0.312 +epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3527 (1.3771) acc 96.8750 (95.6250) lr 0.260000 +epoch: [36/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:21:51 loss 1.5210 (1.4249) acc 90.6250 (94.6094) lr 0.260000 +FPS@all 820.347, TIME@all 0.312 +epoch: [37/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:21:48 loss 1.2678 (1.3994) acc 100.0000 (95.6250) lr 0.260000 +epoch: [37/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 1:21:41 loss 1.4121 (1.4427) acc 90.6250 (94.3750) lr 0.260000 +FPS@all 819.629, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4702 (1.3943) acc 90.6250 (95.1562) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.4820 (1.4468) acc 93.7500 (93.5156) lr 0.260000 +FPS@all 819.499, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5343 (1.4002) acc 90.6250 (95.4688) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.3859 (1.4367) acc 93.7500 (93.9844) lr 0.260000 +FPS@all 819.510, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4441 (1.3728) acc 96.8750 (95.9375) lr 0.260000 +epoch: [37/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.2612 (1.4107) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 819.539, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4820 (1.3657) acc 90.6250 (95.0000) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:21:41 loss 1.4381 (1.4417) acc 93.7500 (93.5938) lr 0.260000 +FPS@all 819.546, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.2983 (1.3554) acc 100.0000 (96.5625) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.6371 (1.4182) acc 90.6250 (95.0000) lr 0.260000 +FPS@all 819.593, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5020 (1.3754) acc 93.7500 (96.2500) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:21:41 loss 1.6395 (1.4411) acc 93.7500 (94.9219) lr 0.260000 +FPS@all 819.533, TIME@all 0.312 +epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5322 (1.4006) acc 96.8750 (96.4062) lr 0.260000 +epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.3662 (1.4501) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 819.537, TIME@all 0.312 +epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.6916 (1.4097) acc 84.3750 (94.5312) lr 0.260000 +epoch: [38/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.5016 (1.4408) acc 93.7500 (94.6094) lr 0.260000 +FPS@all 821.207, TIME@all 0.312 +epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 1:21:17 loss 1.5796 (1.3905) acc 90.6250 (94.8438) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:14 loss 1.3329 (1.4465) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 821.259, TIME@all 0.312 +epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.5176 (1.4094) acc 93.7500 (95.4688) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:21:15 loss 1.4438 (1.4662) acc 90.6250 (93.2812) lr 0.260000 +FPS@all 821.113, TIME@all 0.312 +epoch: [38/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:21:18 loss 1.5379 (1.4251) acc 93.7500 (95.6250) lr 0.260000 +epoch: [38/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3989 (1.4589) acc 90.6250 (93.5938) lr 0.260000 +FPS@all 821.108, TIME@all 0.312 +epoch: [38/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 1:21:18 loss 1.4942 (1.3887) acc 90.6250 (95.9375) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3453 (1.4357) acc 100.0000 (94.8438) lr 0.260000 +FPS@all 821.128, TIME@all 0.312 +epoch: [38/350][20/50] time 0.309 (0.312) data 0.001 (0.012) eta 1:21:18 loss 1.5970 (1.4194) acc 90.6250 (95.4688) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:21:16 loss 1.6704 (1.4644) acc 84.3750 (94.0625) lr 0.260000 +FPS@all 821.079, TIME@all 0.312 +epoch: [38/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:21:20 loss 1.6906 (1.4336) acc 87.5000 (95.0000) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3288 (1.4549) acc 100.0000 (94.2188) lr 0.260000 +FPS@all 821.126, TIME@all 0.312 +epoch: [38/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.6247 (1.4269) acc 90.6250 (94.6875) lr 0.260000 +epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.5127 (1.4692) acc 93.7500 (93.7500) lr 0.260000 +FPS@all 821.116, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:21:05 loss 1.7075 (1.4950) acc 84.3750 (93.9062) lr 0.260000 +epoch: [39/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 1:20:55 loss 1.6401 (1.5680) acc 84.3750 (91.6406) lr 0.260000 +FPS@all 821.602, TIME@all 0.312 +epoch: [39/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:21:04 loss 1.5221 (1.4659) acc 96.8750 (94.0625) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:55 loss 1.3779 (1.5115) acc 96.8750 (92.9688) lr 0.260000 +FPS@all 821.572, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.001 (0.012) eta 1:21:06 loss 1.6500 (1.4955) acc 90.6250 (92.8125) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.3543 (1.5335) acc 96.8750 (92.4219) lr 0.260000 +FPS@all 821.531, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:05 loss 1.5484 (1.4817) acc 90.6250 (93.4375) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.5722 (1.5379) acc 90.6250 (91.4062) lr 0.260000 +FPS@all 821.532, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.7963 (1.4872) acc 87.5000 (93.1250) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.3261 (1.5269) acc 96.8750 (92.1875) lr 0.260000 +FPS@all 821.452, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.5366 (1.4792) acc 90.6250 (94.5312) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.4726 (1.5668) acc 96.8750 (91.0938) lr 0.260000 +FPS@all 821.488, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.6423 (1.4807) acc 90.6250 (93.4375) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.5047 (1.5548) acc 84.3750 (91.0938) lr 0.260000 +FPS@all 821.515, TIME@all 0.312 +epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:21:06 loss 1.7129 (1.5195) acc 87.5000 (92.0312) lr 0.260000 +epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:56 loss 1.4877 (1.5402) acc 90.6250 (91.6406) lr 0.260000 +FPS@all 821.499, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3136 (1.4287) acc 96.8750 (94.8438) lr 0.260000 +epoch: [40/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:46 loss 1.5473 (1.4428) acc 87.5000 (94.7656) lr 0.260000 +FPS@all 820.286, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:53 loss 1.5038 (1.4372) acc 96.8750 (94.5312) lr 0.260000 +epoch: [40/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.3181 (1.4643) acc 96.8750 (93.8281) lr 0.260000 +FPS@all 820.091, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3836 (1.3867) acc 96.8750 (96.4062) lr 0.260000 +epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.5694 (1.4233) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 820.184, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:20:52 loss 1.4305 (1.4083) acc 93.7500 (95.7812) lr 0.260000 +epoch: [40/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:46 loss 1.6908 (1.4361) acc 90.6250 (95.5469) lr 0.260000 +FPS@all 820.211, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.2724 (1.4201) acc 100.0000 (96.0938) lr 0.260000 +epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.5743 (1.4472) acc 90.6250 (94.6875) lr 0.260000 +FPS@all 820.126, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:53 loss 1.4177 (1.3952) acc 96.8750 (95.3125) lr 0.260000 +epoch: [40/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:20:47 loss 1.4582 (1.4385) acc 93.7500 (94.1406) lr 0.260000 +FPS@all 820.104, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 1:20:52 loss 1.2783 (1.3927) acc 100.0000 (95.1562) lr 0.260000 +epoch: [40/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:20:47 loss 1.5651 (1.4155) acc 93.7500 (94.8438) lr 0.260000 +FPS@all 820.155, TIME@all 0.312 +epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3661 (1.4290) acc 100.0000 (94.6875) lr 0.260000 +epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.4949 (1.4459) acc 93.7500 (93.8281) lr 0.260000 +FPS@all 820.162, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:41 loss 1.3526 (1.4491) acc 93.7500 (94.2188) lr 0.260000 +epoch: [41/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4507 (1.4588) acc 93.7500 (93.7500) lr 0.260000 +FPS@all 819.526, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:42 loss 1.3630 (1.4536) acc 100.0000 (93.4375) lr 0.260000 +epoch: [41/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4105 (1.4428) acc 96.8750 (93.8281) lr 0.260000 +FPS@all 819.398, TIME@all 0.312 +epoch: [41/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:20:42 loss 1.5720 (1.4376) acc 90.6250 (94.2188) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:37 loss 1.7688 (1.4663) acc 87.5000 (93.8281) lr 0.260000 +FPS@all 819.308, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:43 loss 1.3269 (1.4335) acc 100.0000 (93.7500) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.3750 (1.4518) acc 100.0000 (93.7500) lr 0.260000 +FPS@all 819.384, TIME@all 0.312 +epoch: [41/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 1:20:43 loss 1.3700 (1.4536) acc 93.7500 (93.1250) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:20:36 loss 1.3912 (1.4566) acc 96.8750 (93.7500) lr 0.260000 +FPS@all 819.389, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:20:43 loss 1.4491 (1.4775) acc 90.6250 (93.9062) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:36 loss 1.4302 (1.4702) acc 93.7500 (94.2969) lr 0.260000 +FPS@all 819.350, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:20:42 loss 1.6112 (1.4449) acc 93.7500 (94.8438) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:36 loss 1.4261 (1.4511) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 819.407, TIME@all 0.312 +epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:42 loss 1.6238 (1.4379) acc 87.5000 (93.9062) lr 0.260000 +epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4377 (1.4466) acc 96.8750 (94.2969) lr 0.260000 +FPS@all 819.416, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.3751 (1.3601) acc 96.8750 (96.0938) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:09 loss 1.4044 (1.3822) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 821.579, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:20:10 loss 1.4092 (1.3451) acc 96.8750 (97.3438) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:20:09 loss 1.3773 (1.3802) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.646, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4146 (1.3546) acc 96.8750 (97.3438) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3126 (1.3702) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 821.491, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:20:11 loss 1.5924 (1.3723) acc 96.8750 (96.7188) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:20:10 loss 1.3402 (1.3693) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 821.508, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4699 (1.3500) acc 96.8750 (96.8750) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3655 (1.3715) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 821.484, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:11 loss 1.3910 (1.3554) acc 93.7500 (96.7188) lr 0.260000 +epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3306 (1.3811) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 821.474, TIME@all 0.312 +epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4292 (1.3661) acc 96.8750 (96.7188) lr 0.260000 +epoch: [42/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3528 (1.3659) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 821.552, TIME@all 0.312 +epoch: [42/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:20:12 loss 1.3161 (1.3244) acc 100.0000 (96.8750) lr 0.260000 +epoch: [42/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:20:09 loss 1.3822 (1.3517) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 821.539, TIME@all 0.312 +epoch: [43/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.5720 (1.3442) acc 90.6250 (95.7812) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:06 loss 1.3959 (1.3895) acc 100.0000 (95.0781) lr 0.260000 +FPS@all 820.256, TIME@all 0.312 +epoch: [43/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:20:08 loss 1.4242 (1.3116) acc 96.8750 (97.3438) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:20:06 loss 1.3312 (1.3414) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 820.298, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.5717 (1.3016) acc 90.6250 (97.9688) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4750 (1.3509) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 820.141, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4149 (1.3027) acc 96.8750 (98.1250) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4016 (1.3672) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 820.192, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4476 (1.3231) acc 93.7500 (97.1875) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.7220 (1.3714) acc 87.5000 (95.6250) lr 0.260000 +FPS@all 820.152, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4069 (1.3272) acc 93.7500 (96.4062) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4421 (1.3561) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 820.224, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:20:09 loss 1.3635 (1.2917) acc 96.8750 (98.4375) lr 0.260000 +epoch: [43/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.3052 (1.3484) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 820.192, TIME@all 0.312 +epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:20:09 loss 1.4941 (1.2949) acc 90.6250 (97.6562) lr 0.260000 +epoch: [43/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:20:07 loss 1.4481 (1.3484) acc 87.5000 (96.0156) lr 0.260000 +FPS@all 820.191, TIME@all 0.312 +epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:19:41 loss 1.3156 (1.3437) acc 100.0000 (96.5625) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.6073 (1.3742) acc 90.6250 (96.0938) lr 0.260000 +FPS@all 821.111, TIME@all 0.312 +epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:19:42 loss 1.3358 (1.3525) acc 93.7500 (95.7812) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3337 (1.4017) acc 100.0000 (95.2344) lr 0.260000 +FPS@all 821.031, TIME@all 0.312 +epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:19:41 loss 1.3583 (1.3675) acc 96.8750 (96.7188) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:19:39 loss 1.3322 (1.3969) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 821.155, TIME@all 0.312 +epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:19:42 loss 1.3579 (1.3598) acc 96.8750 (96.4062) lr 0.260000 +epoch: [44/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3952 (1.3889) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 821.018, TIME@all 0.312 +epoch: [44/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:19:42 loss 1.3034 (1.3288) acc 96.8750 (97.5000) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:19:40 loss 1.3271 (1.3669) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 821.036, TIME@all 0.312 +epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:19:41 loss 1.5125 (1.3671) acc 93.7500 (96.5625) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.2448 (1.3999) acc 100.0000 (95.4688) lr 0.260000 +FPS@all 821.086, TIME@all 0.312 +epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:19:41 loss 1.5351 (1.3678) acc 93.7500 (95.4688) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3193 (1.4079) acc 100.0000 (95.0000) lr 0.260000 +FPS@all 821.089, TIME@all 0.312 +epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:19:42 loss 1.5648 (1.3777) acc 84.3750 (95.7812) lr 0.260000 +epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.4814 (1.3969) acc 93.7500 (95.0000) lr 0.260000 +FPS@all 821.077, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.3028 (1.3831) acc 93.7500 (96.0938) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4174 (1.4084) acc 93.7500 (95.2344) lr 0.260000 +FPS@all 820.384, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.5405 (1.3712) acc 93.7500 (96.0938) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:29 loss 1.5865 (1.4140) acc 90.6250 (94.9219) lr 0.260000 +FPS@all 820.419, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:40 loss 1.3798 (1.3600) acc 96.8750 (96.5625) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.5476 (1.3849) acc 90.6250 (95.7031) lr 0.260000 +FPS@all 820.256, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.6148 (1.3605) acc 90.6250 (95.1562) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4632 (1.4001) acc 87.5000 (94.4531) lr 0.260000 +FPS@all 820.325, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:19:39 loss 1.2842 (1.3618) acc 100.0000 (96.7188) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:19:30 loss 1.4112 (1.3888) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 820.287, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.4972 (1.3640) acc 93.7500 (96.7188) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4453 (1.4164) acc 90.6250 (95.0000) lr 0.260000 +FPS@all 820.319, TIME@all 0.312 +epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.4299 (1.3404) acc 93.7500 (96.7188) lr 0.260000 +epoch: [45/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4271 (1.3888) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 820.309, TIME@all 0.312 +epoch: [45/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.3147 (1.3426) acc 90.6250 (96.8750) lr 0.260000 +epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.2999 (1.3731) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 820.342, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:19:24 loss 1.6308 (1.4030) acc 87.5000 (95.6250) lr 0.260000 +epoch: [46/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.5723 (1.4542) acc 84.3750 (93.8281) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.805, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.6638 (1.4344) acc 90.6250 (93.5938) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:06 loss 1.3694 (1.4628) acc 96.8750 (93.4375) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.866, TIME@all 0.311 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.6148 (1.4060) acc 84.3750 (95.0000) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.3798 (1.4765) acc 93.7500 (92.8906) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.718, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.5522 (1.3960) acc 87.5000 (95.3125) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.5152 (1.4353) acc 96.8750 (94.4531) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.789, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:19:24 loss 1.4964 (1.4024) acc 87.5000 (95.4688) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:07 loss 1.4005 (1.4414) acc 93.7500 (93.9062) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.795, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:25 loss 1.6603 (1.4177) acc 90.6250 (94.3750) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.4840 (1.4497) acc 87.5000 (93.6719) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.744, TIME@all 0.312 +epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.5089 (1.3875) acc 90.6250 (95.1562) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:07 loss 1.3656 (1.4401) acc 100.0000 (93.9844) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.791, TIME@all 0.312 +epoch: [46/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:19:25 loss 1.5450 (1.4016) acc 96.8750 (95.7812) lr 0.260000 +epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.4670 (1.4314) acc 93.7500 (94.8438) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.783, TIME@all 0.312 +epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:19:08 loss 1.5782 (1.3510) acc 84.3750 (96.7188) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:18:59 loss 1.3741 (1.3982) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 820.125, TIME@all 0.312 +epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.6272 (1.3306) acc 90.6250 (97.3438) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:18:59 loss 1.3923 (1.3921) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 820.176, TIME@all 0.312 +epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.4480 (1.3635) acc 93.7500 (95.6250) lr 0.260000 +epoch: [47/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.3510 (1.4005) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 820.047, TIME@all 0.312 +epoch: [47/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.5146 (1.3461) acc 90.6250 (97.0312) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.4076 (1.3924) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 820.117, TIME@all 0.312 +epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:19:09 loss 1.4206 (1.3304) acc 96.8750 (97.3438) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:19:00 loss 1.4131 (1.3667) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 820.062, TIME@all 0.312 +epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:08 loss 1.5562 (1.3626) acc 93.7500 (96.2500) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:18:59 loss 1.2557 (1.4036) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 820.111, TIME@all 0.312 +epoch: [47/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:10 loss 1.5224 (1.3827) acc 90.6250 (95.7812) lr 0.260000 +epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.3557 (1.4102) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 820.075, TIME@all 0.312 +epoch: [47/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.4529 (1.3412) acc 90.6250 (96.2500) lr 0.260000 +epoch: [47/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:19:00 loss 1.3458 (1.3874) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 820.034, TIME@all 0.312 +epoch: [48/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:18:51 loss 1.5871 (1.3541) acc 93.7500 (96.2500) lr 0.260000 +epoch: [48/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.5022 (1.4011) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 819.596, TIME@all 0.312 +epoch: [48/350][20/50] time 0.314 (0.313) data 0.001 (0.014) eta 1:18:50 loss 1.3973 (1.3341) acc 93.7500 (97.0312) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.3480 (1.3710) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 819.626, TIME@all 0.312 +epoch: [48/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.2933 (1.3441) acc 100.0000 (96.5625) lr 0.260000 +epoch: [48/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:18:52 loss 1.2812 (1.3953) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 819.497, TIME@all 0.312 +epoch: [48/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:18:50 loss 1.3780 (1.3599) acc 96.8750 (96.4062) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.5490 (1.4120) acc 90.6250 (94.6094) lr 0.260000 +FPS@all 819.515, TIME@all 0.312 +epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.2684 (1.3071) acc 100.0000 (97.8125) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.4388 (1.3687) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 819.521, TIME@all 0.312 +epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.3539 (1.3466) acc 100.0000 (96.7188) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.3005 (1.3813) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 819.493, TIME@all 0.312 +epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.6349 (1.3714) acc 90.6250 (95.4688) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.4643 (1.3969) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 819.532, TIME@all 0.312 +epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.4287 (1.3238) acc 96.8750 (97.5000) lr 0.260000 +epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.3902 (1.3950) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 819.485, TIME@all 0.312 +epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:34 loss 1.4676 (1.3558) acc 96.8750 (96.5625) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:25 loss 1.5682 (1.3841) acc 90.6250 (95.7812) lr 0.260000 +FPS@all 820.362, TIME@all 0.312 +epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.5353 (1.3803) acc 90.6250 (96.5625) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:26 loss 1.5654 (1.4075) acc 87.5000 (94.9219) lr 0.260000 +FPS@all 820.195, TIME@all 0.312 +epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:18:35 loss 1.4409 (1.3654) acc 93.7500 (96.4062) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:18:26 loss 1.4084 (1.3881) acc 93.7500 (95.4688) lr 0.260000 +FPS@all 820.185, TIME@all 0.312 +epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:36 loss 1.4464 (1.3442) acc 90.6250 (95.6250) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:18:26 loss 1.4018 (1.3672) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 820.166, TIME@all 0.312 +epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.3185 (1.3620) acc 96.8750 (95.7812) lr 0.260000 +epoch: [49/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:18:26 loss 1.4061 (1.4047) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 820.220, TIME@all 0.312 +epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.4886 (1.3783) acc 93.7500 (95.7812) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 1:18:26 loss 1.6209 (1.4216) acc 84.3750 (94.2188) lr 0.260000 +FPS@all 820.204, TIME@all 0.312 +epoch: [49/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:18:35 loss 1.5025 (1.3552) acc 96.8750 (97.1875) lr 0.260000 +epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:26 loss 1.5574 (1.3767) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 820.222, TIME@all 0.312 +epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.3717 (1.3555) acc 93.7500 (95.1562) lr 0.260000 +epoch: [49/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 1:18:26 loss 1.6093 (1.3936) acc 90.6250 (95.0781) lr 0.260000 +FPS@all 820.244, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:22 loss 1.3800 (1.3714) acc 96.8750 (96.2500) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.3501 (1.4032) acc 93.7500 (95.2344) lr 0.260000 +FPS@all 819.511, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:18:22 loss 1.3873 (1.4031) acc 96.8750 (94.8438) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:18:17 loss 1.3209 (1.4104) acc 96.8750 (94.5312) lr 0.260000 +FPS@all 819.548, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.4938 (1.3871) acc 93.7500 (96.5625) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.5903 (1.4055) acc 87.5000 (95.0781) lr 0.260000 +FPS@all 819.413, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.6864 (1.4180) acc 78.1250 (94.5312) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.3742 (1.4239) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 819.470, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.3207 (1.3637) acc 96.8750 (96.5625) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.4403 (1.4050) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 819.456, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:18:23 loss 1.4274 (1.3631) acc 93.7500 (96.5625) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.2475 (1.4009) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 819.437, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:22 loss 1.5662 (1.4190) acc 93.7500 (95.3125) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:17 loss 1.3772 (1.4233) acc 96.8750 (94.9219) lr 0.260000 +FPS@all 819.533, TIME@all 0.312 +epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.4062 (1.3982) acc 96.8750 (95.1562) lr 0.260000 +epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.2962 (1.3979) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 819.467, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:17:58 loss 1.3187 (1.3741) acc 100.0000 (95.0000) lr 0.260000 +epoch: [51/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:52 loss 1.4037 (1.4024) acc 90.6250 (93.8281) lr 0.260000 +FPS@all 820.915, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.3569 (1.3693) acc 93.7500 (96.0938) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.5966 (1.3887) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 820.799, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.3892 (1.3467) acc 96.8750 (96.8750) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.5123 (1.3934) acc 87.5000 (95.2344) lr 0.260000 +FPS@all 820.728, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:18:00 loss 1.5041 (1.3736) acc 87.5000 (95.9375) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4508 (1.4052) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 820.738, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4019 (1.3556) acc 93.7500 (96.8750) lr 0.260000 +epoch: [51/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 1:17:53 loss 1.3576 (1.3828) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 820.809, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:17:59 loss 1.5000 (1.3812) acc 90.6250 (95.6250) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 1:17:53 loss 1.4261 (1.4013) acc 93.7500 (94.9219) lr 0.260000 +FPS@all 820.771, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4435 (1.3980) acc 100.0000 (95.4688) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4925 (1.4054) acc 93.7500 (94.7656) lr 0.260000 +FPS@all 820.791, TIME@all 0.312 +epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4818 (1.3598) acc 93.7500 (96.2500) lr 0.260000 +epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4352 (1.3777) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 820.796, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.4987 (1.3678) acc 93.7500 (95.6250) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4741 (1.4008) acc 90.6250 (94.9219) lr 0.260000 +FPS@all 820.347, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 1:17:48 loss 1.5134 (1.3559) acc 90.6250 (95.7812) lr 0.260000 +epoch: [52/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4777 (1.3850) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 820.366, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.4166 (1.3802) acc 93.7500 (95.4688) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:42 loss 1.3828 (1.3914) acc 100.0000 (95.0000) lr 0.260000 +FPS@all 820.228, TIME@all 0.312 +epoch: [52/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.5333 (1.3776) acc 87.5000 (96.2500) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:42 loss 1.4610 (1.3874) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 820.214, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.3774 (1.3642) acc 96.8750 (96.2500) lr 0.260000 +epoch: [52/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4123 (1.3930) acc 96.8750 (95.0000) lr 0.260000 +FPS@all 820.274, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.4541 (1.3444) acc 93.7500 (96.8750) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4111 (1.3733) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 820.310, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.3968 (1.3486) acc 96.8750 (96.4062) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:41 loss 1.3575 (1.3754) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 820.306, TIME@all 0.312 +epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:48 loss 1.3021 (1.3754) acc 100.0000 (95.7812) lr 0.260000 +epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:41 loss 1.4398 (1.4069) acc 96.8750 (95.0781) lr 0.260000 +FPS@all 820.281, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:17:32 loss 1.2616 (1.3148) acc 100.0000 (97.1875) lr 0.260000 +epoch: [53/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:17:22 loss 1.4549 (1.3538) acc 90.6250 (96.1719) lr 0.260000 +FPS@all 821.172, TIME@all 0.312 +epoch: [53/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:17:31 loss 1.3494 (1.3551) acc 96.8750 (96.2500) lr 0.260000 +epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:17:22 loss 1.4122 (1.3669) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 821.243, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:32 loss 1.2466 (1.3352) acc 100.0000 (97.5000) lr 0.260000 +epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:22 loss 1.4051 (1.3527) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.132, TIME@all 0.312 +epoch: [53/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2417 (1.3178) acc 100.0000 (97.3438) lr 0.260000 +epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:23 loss 1.4925 (1.3329) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 821.093, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2761 (1.3295) acc 100.0000 (96.4062) lr 0.260000 +epoch: [53/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.5802 (1.3603) acc 87.5000 (96.0938) lr 0.260000 +FPS@all 821.197, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.3427 (1.3480) acc 96.8750 (96.7188) lr 0.260000 +epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:22 loss 1.3227 (1.3739) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 821.171, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:32 loss 1.3115 (1.3203) acc 96.8750 (96.7188) lr 0.260000 +epoch: [53/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.4330 (1.3624) acc 90.6250 (95.7031) lr 0.260000 +FPS@all 821.191, TIME@all 0.312 +epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2378 (1.3290) acc 100.0000 (96.8750) lr 0.260000 +epoch: [53/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.5431 (1.3622) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 821.158, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.3656 (1.3364) acc 93.7500 (97.0312) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.4743 (1.3690) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 821.433, TIME@all 0.312 +epoch: [54/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:17:05 loss 1.3529 (1.3624) acc 100.0000 (96.0938) lr 0.260000 +epoch: [54/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.4001 (1.3907) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 821.498, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4186 (1.3545) acc 90.6250 (95.9375) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:17:02 loss 1.4172 (1.3919) acc 90.6250 (95.5469) lr 0.260000 +FPS@all 821.375, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4387 (1.3183) acc 90.6250 (97.8125) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.2952 (1.3574) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 821.410, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:17:06 loss 1.3667 (1.3277) acc 90.6250 (96.7188) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:17:02 loss 1.2975 (1.3587) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 821.383, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4305 (1.3552) acc 96.8750 (96.0938) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.3572 (1.3876) acc 90.6250 (95.2344) lr 0.260000 +FPS@all 821.414, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.3067 (1.3445) acc 100.0000 (96.7188) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.2166 (1.3887) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 821.378, TIME@all 0.312 +epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4257 (1.3356) acc 93.7500 (97.6562) lr 0.260000 +epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:01 loss 1.3056 (1.3748) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 821.427, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:28 loss 1.4950 (1.3674) acc 93.7500 (96.4062) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.4399 (1.3768) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 819.501, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:17:26 loss 1.6928 (1.3991) acc 87.5000 (95.0000) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 1:17:01 loss 1.5761 (1.3802) acc 90.6250 (95.9375) lr 0.260000 +FPS@all 819.592, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5493 (1.3744) acc 87.5000 (94.8438) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.4327 (1.3879) acc 90.6250 (94.8438) lr 0.260000 +FPS@all 819.396, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.3978 (1.3819) acc 93.7500 (94.8438) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.6188 (1.4103) acc 84.3750 (93.7500) lr 0.260000 +FPS@all 819.422, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5037 (1.3745) acc 93.7500 (95.3125) lr 0.260000 +epoch: [55/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.5257 (1.3956) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 819.499, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5265 (1.3772) acc 93.7500 (95.7812) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.3765 (1.3875) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 819.466, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.7319 (1.3677) acc 84.3750 (95.9375) lr 0.260000 +epoch: [55/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:17:02 loss 1.5526 (1.3971) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 819.442, TIME@all 0.312 +epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.4391 (1.3580) acc 93.7500 (97.1875) lr 0.260000 +epoch: [55/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.4074 (1.3920) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 819.466, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3539 (1.3998) acc 96.8750 (94.6875) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:36 loss 1.4322 (1.4137) acc 93.7500 (94.0625) lr 0.260000 +FPS@all 820.703, TIME@all 0.312 +epoch: [56/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:16:36 loss 1.2442 (1.3675) acc 100.0000 (95.6250) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:16:36 loss 1.3242 (1.3817) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 820.788, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3510 (1.4060) acc 100.0000 (96.0938) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5345 (1.4360) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 820.671, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:36 loss 1.2942 (1.3909) acc 96.8750 (95.3125) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5237 (1.4012) acc 93.7500 (95.8594) lr 0.260000 +FPS@all 820.615, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3778 (1.4088) acc 100.0000 (94.2188) lr 0.260000 +epoch: [56/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5411 (1.4325) acc 93.7500 (93.8281) lr 0.260000 +FPS@all 820.714, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3143 (1.3882) acc 96.8750 (96.4062) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.3789 (1.4017) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 820.641, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3195 (1.3938) acc 96.8750 (95.6250) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.4133 (1.4111) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 820.597, TIME@all 0.312 +epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3093 (1.3846) acc 100.0000 (96.2500) lr 0.260000 +epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:36 loss 1.5012 (1.4213) acc 93.7500 (94.6094) lr 0.260000 +FPS@all 820.703, TIME@all 0.312 +epoch: [57/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:16:39 loss 1.2969 (1.3020) acc 93.7500 (97.5000) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:16:29 loss 1.3856 (1.3631) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 819.374, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:40 loss 1.3358 (1.3336) acc 96.8750 (96.2500) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.4254 (1.3565) acc 87.5000 (95.7812) lr 0.260000 +FPS@all 819.228, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:39 loss 1.2724 (1.3447) acc 100.0000 (96.8750) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:16:29 loss 1.4266 (1.3710) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 819.298, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:39 loss 1.3339 (1.3362) acc 100.0000 (96.5625) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3892 (1.3603) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 819.323, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:40 loss 1.2520 (1.3312) acc 100.0000 (97.8125) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3417 (1.3606) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 819.247, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:39 loss 1.3003 (1.3344) acc 96.8750 (96.2500) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.2387 (1.3505) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 819.266, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:39 loss 1.3837 (1.3162) acc 93.7500 (96.5625) lr 0.260000 +epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3551 (1.3598) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 819.285, TIME@all 0.312 +epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:39 loss 1.3726 (1.3511) acc 96.8750 (96.0938) lr 0.260000 +epoch: [57/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.4399 (1.3749) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 819.279, TIME@all 0.312 +epoch: [58/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:16:15 loss 1.4228 (1.3011) acc 93.7500 (97.5000) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.4076 (1.3190) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 820.565, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:15 loss 1.3550 (1.3135) acc 90.6250 (97.0312) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2265 (1.3365) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 820.468, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:16 loss 1.3048 (1.2954) acc 96.8750 (97.1875) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:06 loss 1.3254 (1.3234) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 820.421, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.3031 (1.3049) acc 96.8750 (97.0312) lr 0.260000 +epoch: [58/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:16:07 loss 1.4554 (1.3353) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.414, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.4405 (1.2731) acc 96.8750 (98.2812) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.3741 (1.3100) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 820.441, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:16:16 loss 1.3387 (1.3032) acc 96.8750 (97.3438) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2582 (1.3156) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 820.434, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:16:15 loss 1.4938 (1.2976) acc 90.6250 (97.5000) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2774 (1.3189) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 820.468, TIME@all 0.312 +epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.3491 (1.3124) acc 93.7500 (97.0312) lr 0.260000 +epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2659 (1.3397) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 820.437, TIME@all 0.312 +epoch: [59/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:49 loss 1.4315 (1.3114) acc 96.8750 (97.1875) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:15:41 loss 1.5558 (1.3501) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 821.663, TIME@all 0.312 +epoch: [59/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:50 loss 1.3540 (1.3244) acc 93.7500 (97.0312) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:15:42 loss 1.4768 (1.3416) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 821.507, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:51 loss 1.5054 (1.3415) acc 90.6250 (97.1875) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:43 loss 1.3163 (1.3432) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 821.474, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.4572 (1.3354) acc 93.7500 (97.3438) lr 0.260000 +epoch: [59/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.4024 (1.3567) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 821.506, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.5054 (1.3294) acc 96.8750 (97.0312) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.2849 (1.3496) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 821.572, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.3944 (1.3224) acc 96.8750 (97.1875) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.3475 (1.3375) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 821.515, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:15:51 loss 1.4910 (1.3231) acc 90.6250 (96.7188) lr 0.260000 +epoch: [59/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 1:15:42 loss 1.4928 (1.3446) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 821.520, TIME@all 0.312 +epoch: [59/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:15:50 loss 1.4647 (1.3047) acc 93.7500 (97.0312) lr 0.260000 +epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.2883 (1.3322) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 821.575, TIME@all 0.312 +epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.3230 (1.4094) acc 96.8750 (95.7812) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5239 (1.4930) acc 87.5000 (92.5000) lr 0.260000 +FPS@all 819.908, TIME@all 0.312 +epoch: [60/350][20/50] time 0.317 (0.313) data 0.000 (0.014) eta 1:15:45 loss 1.2973 (1.4172) acc 100.0000 (93.9062) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.4208 (1.4904) acc 93.7500 (92.4219) lr 0.260000 +FPS@all 819.960, TIME@all 0.312 +epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.6219 (1.3985) acc 90.6250 (94.0625) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5348 (1.4697) acc 96.8750 (93.5938) lr 0.260000 +FPS@all 819.843, TIME@all 0.312 +epoch: [60/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:15:46 loss 1.3096 (1.4152) acc 100.0000 (95.6250) lr 0.260000 +epoch: [60/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:15:37 loss 1.4139 (1.4823) acc 93.7500 (93.2031) lr 0.260000 +FPS@all 819.836, TIME@all 0.312 +epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.5059 (1.4003) acc 93.7500 (95.7812) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 1:15:37 loss 1.2985 (1.4768) acc 96.8750 (93.8281) lr 0.260000 +FPS@all 819.878, TIME@all 0.312 +epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:47 loss 1.4281 (1.3668) acc 100.0000 (96.7188) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 1:15:37 loss 1.5582 (1.4508) acc 93.7500 (93.8281) lr 0.260000 +FPS@all 819.832, TIME@all 0.312 +epoch: [60/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 1:15:46 loss 1.3592 (1.3966) acc 93.7500 (95.1562) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5141 (1.4944) acc 96.8750 (93.2031) lr 0.260000 +FPS@all 819.899, TIME@all 0.312 +epoch: [60/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:15:46 loss 1.4152 (1.4087) acc 96.8750 (95.3125) lr 0.260000 +epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.3910 (1.4835) acc 93.7500 (93.5156) lr 0.260000 +FPS@all 819.879, TIME@all 0.312 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:15:17 loss 1.3886 (1.3982) acc 93.7500 (96.0938) lr 0.260000 +epoch: [61/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:15:06 loss 1.3181 (1.4085) acc 100.0000 (95.5469) lr 0.260000 +FPS@all 822.647, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.2945 (1.3787) acc 96.8750 (95.7812) lr 0.260000 +epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.2840 (1.3854) acc 96.8750 (95.3906) lr 0.260000 +FPS@all 822.557, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3210 (1.3745) acc 100.0000 (95.4688) lr 0.260000 +epoch: [61/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.3871 (1.4055) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 822.491, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3167 (1.3628) acc 100.0000 (96.4062) lr 0.260000 +epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.3508 (1.3855) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 822.462, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.4323 (1.3670) acc 93.7500 (96.7188) lr 0.260000 +epoch: [61/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.2593 (1.3810) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 822.475, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.4000 (1.3533) acc 90.6250 (95.6250) lr 0.260000 +epoch: [61/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.3624 (1.3863) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 822.523, TIME@all 0.311 +epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3816 (1.3608) acc 100.0000 (96.7188) lr 0.260000 +epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.3090 (1.3860) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 822.534, TIME@all 0.311 +epoch: [61/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3386 (1.3801) acc 100.0000 (96.4062) lr 0.260000 +epoch: [61/350][40/50] time 0.308 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.2915 (1.4133) acc 100.0000 (94.6875) lr 0.260000 +FPS@all 822.502, TIME@all 0.311 +epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.4854 (1.4114) acc 93.7500 (95.3125) lr 0.260000 +epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3892 (1.4241) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 820.512, TIME@all 0.312 +epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.5050 (1.3616) acc 93.7500 (97.1875) lr 0.260000 +epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3270 (1.4067) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 820.553, TIME@all 0.312 +epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.8750 (1.3964) acc 84.3750 (95.9375) lr 0.260000 +epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.4430 (1.4283) acc 90.6250 (94.5312) lr 0.260000 +FPS@all 820.408, TIME@all 0.312 +epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:15:08 loss 1.6982 (1.4272) acc 90.6250 (94.0625) lr 0.260000 +epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:15:03 loss 1.4117 (1.4331) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 820.373, TIME@all 0.312 +epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.7935 (1.4276) acc 87.5000 (95.3125) lr 0.260000 +epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3242 (1.4547) acc 96.8750 (93.9062) lr 0.260000 +FPS@all 820.492, TIME@all 0.312 +epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.5440 (1.4004) acc 90.6250 (95.4688) lr 0.260000 +epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.4107 (1.4412) acc 96.8750 (94.1406) lr 0.260000 +FPS@all 820.450, TIME@all 0.312 +epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.6733 (1.3547) acc 93.7500 (96.7188) lr 0.260000 +epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3298 (1.4043) acc 100.0000 (95.3125) lr 0.260000 +FPS@all 820.443, TIME@all 0.312 +epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.6377 (1.4091) acc 87.5000 (95.1562) lr 0.260000 +epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3386 (1.4293) acc 96.8750 (94.9219) lr 0.260000 +FPS@all 820.440, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:45 loss 1.3257 (1.3702) acc 96.8750 (95.3125) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:49 loss 1.4062 (1.4044) acc 96.8750 (94.6094) lr 0.260000 +FPS@all 820.262, TIME@all 0.312 +epoch: [63/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.3629 (1.3564) acc 96.8750 (95.7812) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:49 loss 1.3849 (1.3915) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 820.203, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3458 (1.3605) acc 96.8750 (96.2500) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4538 (1.4029) acc 93.7500 (94.8438) lr 0.260000 +FPS@all 820.081, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.011) eta 1:14:48 loss 1.3289 (1.3557) acc 93.7500 (96.0938) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3478 (1.4002) acc 100.0000 (94.7656) lr 0.260000 +FPS@all 820.054, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.4744 (1.3672) acc 96.8750 (97.0312) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4213 (1.4026) acc 90.6250 (95.7812) lr 0.260000 +FPS@all 820.079, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.5316 (1.3712) acc 90.6250 (95.4688) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3408 (1.4130) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 820.111, TIME@all 0.312 +epoch: [63/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3519 (1.3766) acc 96.8750 (96.4062) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4970 (1.4243) acc 87.5000 (94.7656) lr 0.260000 +FPS@all 820.087, TIME@all 0.312 +epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3891 (1.3664) acc 96.8750 (96.7188) lr 0.260000 +epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3838 (1.3908) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 820.113, TIME@all 0.312 +epoch: [64/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:14:33 loss 1.3760 (1.3813) acc 96.8750 (95.3125) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3408 (1.4015) acc 96.8750 (94.8438) lr 0.260000 +FPS@all 821.066, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:33 loss 1.6249 (1.3906) acc 93.7500 (95.7812) lr 0.260000 +epoch: [64/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.2842 (1.3759) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 820.968, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:14:34 loss 1.6213 (1.3856) acc 90.6250 (96.8750) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:14:30 loss 1.2972 (1.3946) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 820.944, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.5978 (1.4006) acc 90.6250 (94.8438) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.001 (0.007) eta 1:14:30 loss 1.3743 (1.4025) acc 96.8750 (94.7656) lr 0.260000 +FPS@all 820.993, TIME@all 0.312 +epoch: [64/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.6688 (1.4140) acc 90.6250 (95.1562) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.4291 (1.4155) acc 93.7500 (94.6094) lr 0.260000 +FPS@all 820.990, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:14:33 loss 1.6619 (1.3685) acc 93.7500 (96.5625) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3428 (1.3931) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 820.983, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.4863 (1.4078) acc 90.6250 (94.5312) lr 0.260000 +epoch: [64/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3527 (1.4146) acc 93.7500 (94.2188) lr 0.260000 +FPS@all 820.992, TIME@all 0.312 +epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:33 loss 1.5091 (1.3952) acc 96.8750 (95.4688) lr 0.260000 +epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:14:30 loss 1.3842 (1.4041) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 821.028, TIME@all 0.312 +epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.5787 (1.3227) acc 90.6250 (97.1875) lr 0.260000 +epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.3743 (1.3373) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 820.783, TIME@all 0.312 +epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.4280 (1.3282) acc 93.7500 (97.0312) lr 0.260000 +epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.3729 (1.3498) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 820.826, TIME@all 0.312 +epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:14:21 loss 1.6557 (1.3269) acc 87.5000 (97.0312) lr 0.260000 +epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3531 (1.3538) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 820.655, TIME@all 0.312 +epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.011) eta 1:14:20 loss 1.6503 (1.3140) acc 90.6250 (97.5000) lr 0.260000 +epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3576 (1.3399) acc 90.6250 (96.7969) lr 0.260000 +FPS@all 820.686, TIME@all 0.312 +epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.4944 (1.3125) acc 90.6250 (97.0312) lr 0.260000 +epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.4898 (1.3406) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 820.710, TIME@all 0.312 +epoch: [65/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:14:20 loss 1.5958 (1.3139) acc 90.6250 (97.6562) lr 0.260000 +epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.4005 (1.3426) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 820.760, TIME@all 0.312 +epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.3709 (1.3300) acc 96.8750 (97.8125) lr 0.260000 +epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.2913 (1.3621) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 820.735, TIME@all 0.312 +epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.5333 (1.3163) acc 90.6250 (97.1875) lr 0.260000 +epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3640 (1.3288) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 820.719, TIME@all 0.312 +epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:14:03 loss 1.3526 (1.3230) acc 96.8750 (96.2500) lr 0.260000 +epoch: [66/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:13:58 loss 1.3208 (1.3400) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 820.786, TIME@all 0.312 +epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.4861 (1.3059) acc 96.8750 (97.8125) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:58 loss 1.3799 (1.3316) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 820.693, TIME@all 0.312 +epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3385 (1.2865) acc 96.8750 (97.6562) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.3671 (1.3243) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 820.663, TIME@all 0.312 +epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3083 (1.2811) acc 100.0000 (97.3438) lr 0.260000 +epoch: [66/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.1961 (1.3205) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 820.623, TIME@all 0.312 +epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3332 (1.2962) acc 96.8750 (97.6562) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.2972 (1.3417) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 820.673, TIME@all 0.312 +epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:14:03 loss 1.4675 (1.2729) acc 93.7500 (98.2812) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.4023 (1.3139) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 820.631, TIME@all 0.312 +epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.4543 (1.2845) acc 93.7500 (97.9688) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:58 loss 1.2716 (1.3108) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 820.712, TIME@all 0.312 +epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3340 (1.2919) acc 100.0000 (97.8125) lr 0.260000 +epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.3022 (1.3185) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 820.680, TIME@all 0.312 +epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.3169 (1.2971) acc 96.8750 (97.6562) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.007) eta 1:14:31 loss 1.2926 (1.3145) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 808.523, TIME@all 0.317 +epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:47 loss 1.3945 (1.2867) acc 90.6250 (97.3438) lr 0.260000 +epoch: [67/350][40/50] time 0.323 (0.316) data 0.000 (0.007) eta 1:14:30 loss 1.3082 (1.3067) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 808.602, TIME@all 0.317 +epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.5320 (1.3086) acc 90.6250 (97.3438) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.007) eta 1:14:31 loss 1.4084 (1.3465) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 808.483, TIME@all 0.317 +epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:13:48 loss 1.5711 (1.3138) acc 90.6250 (97.0312) lr 0.260000 +epoch: [67/350][40/50] time 0.325 (0.316) data 0.000 (0.006) eta 1:14:31 loss 1.3345 (1.3450) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 808.459, TIME@all 0.317 +epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4124 (1.2944) acc 90.6250 (96.8750) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.4156 (1.3483) acc 90.6250 (95.6250) lr 0.260000 +FPS@all 808.550, TIME@all 0.317 +epoch: [67/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:13:50 loss 1.4938 (1.3460) acc 93.7500 (97.0312) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.006) eta 1:14:31 loss 1.4156 (1.3471) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 808.462, TIME@all 0.317 +epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4914 (1.3259) acc 90.6250 (96.7188) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.3442 (1.3558) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 808.507, TIME@all 0.317 +epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4994 (1.3031) acc 93.7500 (97.3438) lr 0.260000 +epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.3067 (1.3418) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 808.473, TIME@all 0.317 +epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 1:13:53 loss 1.3202 (1.3111) acc 93.7500 (97.0312) lr 0.260000 +epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3929 (1.3492) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 817.806, TIME@all 0.313 +epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.015) eta 1:13:54 loss 1.2785 (1.3081) acc 100.0000 (97.6562) lr 0.260000 +epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3255 (1.3416) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 817.701, TIME@all 0.313 +epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.4153 (1.3334) acc 96.8750 (97.3438) lr 0.260000 +epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:40 loss 1.6029 (1.3578) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 817.602, TIME@all 0.313 +epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:55 loss 1.3260 (1.3027) acc 96.8750 (97.3438) lr 0.260000 +epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:40 loss 1.2901 (1.3256) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 817.560, TIME@all 0.313 +epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.2576 (1.3241) acc 100.0000 (97.1875) lr 0.260000 +epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3327 (1.3721) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 817.644, TIME@all 0.313 +epoch: [68/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.3895 (1.3108) acc 90.6250 (97.5000) lr 0.260000 +epoch: [68/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3920 (1.3487) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 817.658, TIME@all 0.313 +epoch: [68/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 1:13:54 loss 1.3293 (1.3118) acc 96.8750 (97.0312) lr 0.260000 +epoch: [68/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:13:38 loss 1.2968 (1.3500) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 817.829, TIME@all 0.313 +epoch: [68/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 1:13:54 loss 1.3395 (1.3010) acc 96.8750 (97.3438) lr 0.260000 +epoch: [68/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:13:39 loss 1.5232 (1.3455) acc 87.5000 (95.9375) lr 0.260000 +FPS@all 817.676, TIME@all 0.313 +epoch: [69/350][20/50] time 0.331 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.3068 (1.2778) acc 96.8750 (98.1250) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.007) eta 1:15:20 loss 1.3625 (1.3282) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 799.432, TIME@all 0.320 +epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.2824 (1.2974) acc 96.8750 (97.0312) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.007) eta 1:15:19 loss 1.3009 (1.3352) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 799.510, TIME@all 0.320 +epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:49 loss 1.2596 (1.2939) acc 100.0000 (97.5000) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.001 (0.006) eta 1:15:20 loss 1.4113 (1.3173) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 799.494, TIME@all 0.320 +epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:48 loss 1.6438 (1.3122) acc 90.6250 (97.6562) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.006) eta 1:15:19 loss 1.4067 (1.3323) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 799.535, TIME@all 0.320 +epoch: [69/350][20/50] time 0.331 (0.323) data 0.001 (0.012) eta 1:15:48 loss 1.2333 (1.2877) acc 96.8750 (97.3438) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.006) eta 1:15:19 loss 1.3307 (1.3271) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 799.468, TIME@all 0.320 +epoch: [69/350][20/50] time 0.331 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.3899 (1.2996) acc 93.7500 (97.0312) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.322) data 0.000 (0.007) eta 1:15:20 loss 1.4748 (1.3348) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 799.361, TIME@all 0.320 +epoch: [69/350][20/50] time 0.331 (0.323) data 0.001 (0.013) eta 1:15:47 loss 1.3700 (1.3080) acc 93.7500 (96.7188) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.321) data 0.001 (0.007) eta 1:15:19 loss 1.3156 (1.3379) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 799.644, TIME@all 0.320 +epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:49 loss 1.2381 (1.2858) acc 100.0000 (97.8125) lr 0.260000 +epoch: [69/350][40/50] time 0.330 (0.322) data 0.000 (0.006) eta 1:15:20 loss 1.3402 (1.3217) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 799.382, TIME@all 0.320 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.015) eta 1:13:34 loss 1.4462 (1.3021) acc 93.7500 (97.3438) lr 0.260000 +epoch: [70/350][40/50] time 0.353 (0.320) data 0.000 (0.008) eta 1:14:49 loss 1.3160 (1.3217) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 786.875, TIME@all 0.325 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 1:13:33 loss 1.3309 (1.3224) acc 96.8750 (96.8750) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3559 (1.3435) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 786.922, TIME@all 0.325 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:35 loss 1.7236 (1.3377) acc 81.2500 (96.8750) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.321) data 0.000 (0.007) eta 1:14:50 loss 1.3513 (1.3426) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 786.709, TIME@all 0.325 +epoch: [70/350][20/50] time 0.315 (0.315) data 0.000 (0.015) eta 1:13:34 loss 1.4369 (1.3108) acc 96.8750 (96.8750) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.320) data 0.000 (0.008) eta 1:14:49 loss 1.3728 (1.3327) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 786.838, TIME@all 0.325 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:35 loss 1.5045 (1.3816) acc 90.6250 (95.6250) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3218 (1.3748) acc 93.7500 (95.5469) lr 0.260000 +FPS@all 786.906, TIME@all 0.325 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 1:13:34 loss 1.3770 (1.2873) acc 90.6250 (97.9688) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.320) data 0.000 (0.007) eta 1:14:49 loss 1.2967 (1.3308) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 786.829, TIME@all 0.325 +epoch: [70/350][20/50] time 0.315 (0.315) data 0.000 (0.014) eta 1:13:34 loss 1.3902 (1.3132) acc 100.0000 (97.8125) lr 0.260000 +epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3886 (1.3696) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 786.806, TIME@all 0.325 +epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:34 loss 1.6675 (1.3114) acc 90.6250 (97.6562) lr 0.260000 +epoch: [70/350][40/50] time 0.353 (0.320) data 0.001 (0.007) eta 1:14:48 loss 1.2903 (1.3335) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 786.938, TIME@all 0.325 +epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.3201 (1.4088) acc 100.0000 (94.5312) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.007) eta 1:14:59 loss 1.2931 (1.4206) acc 100.0000 (94.5312) lr 0.260000 +FPS@all 801.191, TIME@all 0.320 +epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.4469 (1.3671) acc 93.7500 (96.5625) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3273 (1.4005) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 801.163, TIME@all 0.320 +epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4743 (1.4013) acc 90.6250 (96.0938) lr 0.260000 +epoch: [71/350][40/50] time 0.310 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3460 (1.3991) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 801.042, TIME@all 0.320 +epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.3255 (1.4043) acc 96.8750 (95.0000) lr 0.260000 +epoch: [71/350][40/50] time 0.313 (0.322) data 0.000 (0.006) eta 1:15:01 loss 1.2859 (1.4076) acc 96.8750 (94.6094) lr 0.260000 +FPS@all 801.038, TIME@all 0.320 +epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.3779 (1.4067) acc 93.7500 (95.0000) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3019 (1.4294) acc 90.6250 (93.9062) lr 0.260000 +FPS@all 801.116, TIME@all 0.320 +epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4024 (1.3578) acc 96.8750 (95.4688) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3468 (1.3994) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 801.079, TIME@all 0.320 +epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4404 (1.3964) acc 90.6250 (94.8438) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.001 (0.006) eta 1:15:00 loss 1.3612 (1.4110) acc 93.7500 (94.4531) lr 0.260000 +FPS@all 801.105, TIME@all 0.320 +epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.2574 (1.3658) acc 96.8750 (96.5625) lr 0.260000 +epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.007) eta 1:15:00 loss 1.3688 (1.3938) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 801.004, TIME@all 0.320 +epoch: [72/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.2588 (1.3232) acc 100.0000 (96.8750) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.4749 (1.3516) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 816.886, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.5033 (1.3127) acc 93.7500 (95.7812) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:49 loss 1.3735 (1.3602) acc 96.8750 (95.3906) lr 0.260000 +FPS@all 816.961, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:39 loss 1.3612 (1.3326) acc 96.8750 (95.6250) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.4471 (1.3799) acc 96.8750 (94.6875) lr 0.260000 +FPS@all 816.799, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:39 loss 1.3819 (1.3577) acc 96.8750 (96.0938) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.2847 (1.3753) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 816.781, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:40 loss 1.5533 (1.3534) acc 90.6250 (96.0938) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.3379 (1.3733) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 816.836, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:40 loss 1.3960 (1.3276) acc 96.8750 (97.3438) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.3034 (1.3530) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 816.819, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.4077 (1.3392) acc 96.8750 (97.5000) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.2800 (1.3780) acc 100.0000 (95.5469) lr 0.260000 +FPS@all 816.813, TIME@all 0.313 +epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:40 loss 1.4823 (1.3107) acc 93.7500 (97.3438) lr 0.260000 +epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.4194 (1.3514) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 816.837, TIME@all 0.313 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:12:33 loss 1.7350 (1.3453) acc 81.2500 (96.2500) lr 0.260000 +epoch: [73/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:12:15 loss 1.2931 (1.3663) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 819.689, TIME@all 0.312 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.4518 (1.3165) acc 96.8750 (96.8750) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:15 loss 1.3037 (1.3452) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 819.693, TIME@all 0.312 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.6234 (1.3297) acc 87.5000 (96.7188) lr 0.260000 +epoch: [73/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.3241 (1.3398) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 819.626, TIME@all 0.312 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.5076 (1.3315) acc 96.8750 (95.3125) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4127 (1.3645) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 819.640, TIME@all 0.312 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.5138 (1.3457) acc 93.7500 (96.0938) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4130 (1.3592) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 819.581, TIME@all 0.312 +epoch: [73/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 1:12:36 loss 1.3803 (1.3118) acc 96.8750 (97.3438) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4032 (1.3329) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 819.604, TIME@all 0.312 +epoch: [73/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:12:34 loss 1.4174 (1.3204) acc 93.7500 (96.7188) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.3684 (1.3432) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 819.595, TIME@all 0.312 +epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:34 loss 1.3281 (1.3079) acc 96.8750 (96.8750) lr 0.260000 +epoch: [73/350][40/50] time 0.308 (0.313) data 0.001 (0.006) eta 1:12:16 loss 1.4075 (1.3462) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 819.614, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:12:00 loss 1.2747 (1.2753) acc 100.0000 (98.1250) lr 0.260000 +epoch: [74/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:11:55 loss 1.4681 (1.3161) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 821.081, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:11:59 loss 1.6443 (1.2969) acc 90.6250 (97.1875) lr 0.260000 +epoch: [74/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:55 loss 1.6623 (1.3288) acc 87.5000 (96.9531) lr 0.260000 +FPS@all 821.139, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.3639 (1.2826) acc 93.7500 (97.6562) lr 0.260000 +epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.6840 (1.3336) acc 87.5000 (96.4844) lr 0.260000 +FPS@all 820.945, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5387 (1.3047) acc 90.6250 (97.0312) lr 0.260000 +epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.5470 (1.3249) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 821.017, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:12:00 loss 1.4990 (1.3065) acc 93.7500 (96.7188) lr 0.260000 +epoch: [74/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:11:55 loss 1.8743 (1.3374) acc 84.3750 (95.7031) lr 0.260000 +FPS@all 820.999, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5047 (1.2917) acc 87.5000 (96.8750) lr 0.260000 +epoch: [74/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.5114 (1.3189) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.967, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5710 (1.3059) acc 90.6250 (97.1875) lr 0.260000 +epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.8039 (1.3411) acc 81.2500 (96.0156) lr 0.260000 +FPS@all 821.005, TIME@all 0.312 +epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:12:00 loss 1.5257 (1.3174) acc 96.8750 (97.3438) lr 0.260000 +epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:11:55 loss 1.5912 (1.3394) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 820.996, TIME@all 0.312 +epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.3714 (1.2960) acc 90.6250 (97.1875) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:45 loss 1.3536 (1.3504) acc 90.6250 (95.7812) lr 0.260000 +FPS@all 818.576, TIME@all 0.313 +epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:11:33 loss 1.5403 (1.2937) acc 90.6250 (96.8750) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:11:45 loss 1.5710 (1.3428) acc 90.6250 (95.9375) lr 0.260000 +FPS@all 818.660, TIME@all 0.313 +epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.5398 (1.3073) acc 87.5000 (97.3438) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3798 (1.3452) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 818.527, TIME@all 0.313 +epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4471 (1.2999) acc 93.7500 (97.9688) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.4409 (1.3435) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 818.540, TIME@all 0.313 +epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.3133 (1.2976) acc 100.0000 (97.5000) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3403 (1.3209) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 818.524, TIME@all 0.313 +epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4362 (1.2806) acc 96.8750 (98.1250) lr 0.260000 +epoch: [75/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3951 (1.3256) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 818.503, TIME@all 0.313 +epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4374 (1.3001) acc 93.7500 (97.1875) lr 0.260000 +epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.4888 (1.3408) acc 90.6250 (96.1719) lr 0.260000 +FPS@all 818.560, TIME@all 0.313 +epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4024 (1.2788) acc 93.7500 (98.9062) lr 0.260000 +epoch: [75/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.5049 (1.3392) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 818.558, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.014) eta 1:12:07 loss 1.3519 (1.2993) acc 96.8750 (97.5000) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3004 (1.3198) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 818.349, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.014) eta 1:12:07 loss 1.4395 (1.3245) acc 90.6250 (96.5625) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.2526 (1.3416) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 818.386, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 1:12:07 loss 1.3187 (1.3020) acc 96.8750 (97.0312) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 1:11:41 loss 1.2950 (1.3354) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 818.232, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.2157 (1.3212) acc 100.0000 (97.1875) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:42 loss 1.2890 (1.3201) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 818.224, TIME@all 0.313 +epoch: [76/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.4337 (1.3021) acc 93.7500 (98.1250) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3319 (1.3292) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 818.301, TIME@all 0.313 +epoch: [76/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.3432 (1.2976) acc 100.0000 (98.2812) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3288 (1.3232) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 818.297, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 1:12:07 loss 1.3679 (1.2907) acc 96.8750 (97.6562) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3178 (1.3210) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 818.283, TIME@all 0.313 +epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.4550 (1.2932) acc 96.8750 (97.9688) lr 0.260000 +epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.2901 (1.3204) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 818.261, TIME@all 0.313 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:10 loss 1.2490 (1.3265) acc 100.0000 (96.5625) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.2620 (1.3544) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 820.386, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:11:10 loss 1.2469 (1.3218) acc 100.0000 (96.0938) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:06 loss 1.3790 (1.3493) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 820.444, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:11:11 loss 1.4093 (1.3142) acc 90.6250 (97.1875) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:11:07 loss 1.3976 (1.3742) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 820.326, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:11 loss 1.2487 (1.3332) acc 100.0000 (96.0938) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5181 (1.3730) acc 93.7500 (95.1562) lr 0.260000 +FPS@all 820.316, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:11:11 loss 1.2383 (1.3242) acc 100.0000 (96.7188) lr 0.260000 +epoch: [77/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.3335 (1.3550) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 820.371, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:10 loss 1.2313 (1.3241) acc 100.0000 (96.4062) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5053 (1.3784) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 820.363, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.001 (0.014) eta 1:11:11 loss 1.4102 (1.3536) acc 96.8750 (95.9375) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5342 (1.3953) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 820.345, TIME@all 0.312 +epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:11 loss 1.3066 (1.3256) acc 100.0000 (96.7188) lr 0.260000 +epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.4554 (1.3576) acc 87.5000 (96.0156) lr 0.260000 +FPS@all 820.329, TIME@all 0.312 +epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4231 (1.3314) acc 93.7500 (96.8750) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:44 loss 1.2819 (1.3486) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 822.262, TIME@all 0.311 +epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3267 (1.3088) acc 96.8750 (97.9688) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3612 (1.3462) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 822.082, TIME@all 0.311 +epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3472 (1.3042) acc 93.7500 (97.1875) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3539 (1.3472) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 822.173, TIME@all 0.311 +epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.011) eta 1:10:44 loss 1.3173 (1.3172) acc 96.8750 (96.8750) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3702 (1.3355) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 822.087, TIME@all 0.311 +epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3286 (1.3147) acc 93.7500 (97.5000) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.2921 (1.3360) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 822.112, TIME@all 0.311 +epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4249 (1.3387) acc 93.7500 (97.0312) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3134 (1.3559) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 822.137, TIME@all 0.311 +epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4161 (1.3064) acc 93.7500 (96.2500) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.4013 (1.3292) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 822.149, TIME@all 0.311 +epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3799 (1.3239) acc 96.8750 (97.0312) lr 0.260000 +epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3503 (1.3431) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 822.164, TIME@all 0.311 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.4074 (1.2914) acc 96.8750 (98.4375) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.2485 (1.3078) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 821.302, TIME@all 0.312 +epoch: [79/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:10:35 loss 1.3860 (1.2744) acc 100.0000 (98.4375) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:10:35 loss 1.2457 (1.2990) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 821.353, TIME@all 0.312 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 1:10:34 loss 1.3618 (1.2760) acc 93.7500 (97.8125) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 1:10:34 loss 1.4623 (1.3079) acc 90.6250 (97.1875) lr 0.260000 +FPS@all 821.505, TIME@all 0.312 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.4612 (1.2922) acc 96.8750 (97.8125) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:36 loss 1.5459 (1.3008) acc 90.6250 (97.1875) lr 0.260000 +FPS@all 821.182, TIME@all 0.312 +epoch: [79/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.2675 (1.2922) acc 96.8750 (97.9688) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.2127 (1.3055) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 821.308, TIME@all 0.312 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.2224 (1.2811) acc 96.8750 (97.9688) lr 0.260000 +epoch: [79/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.3211 (1.3100) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 821.292, TIME@all 0.312 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:35 loss 1.4944 (1.3108) acc 87.5000 (96.5625) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:10:35 loss 1.3301 (1.3139) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.233, TIME@all 0.312 +epoch: [79/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 1:10:34 loss 1.4648 (1.2955) acc 96.8750 (96.7188) lr 0.260000 +epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.4429 (1.3061) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 821.299, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.2672 (1.3116) acc 93.7500 (97.0312) lr 0.260000 +epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 1:10:27 loss 1.3655 (1.3390) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 819.871, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.2460 (1.2896) acc 100.0000 (97.9688) lr 0.260000 +epoch: [80/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:10:26 loss 1.2937 (1.3242) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 819.970, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:33 loss 1.2015 (1.2893) acc 100.0000 (98.2812) lr 0.260000 +epoch: [80/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2382 (1.3354) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 819.796, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:34 loss 1.2194 (1.2836) acc 100.0000 (98.1250) lr 0.260000 +epoch: [80/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2275 (1.3211) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 819.820, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:34 loss 1.2786 (1.3173) acc 96.8750 (96.4062) lr 0.260000 +epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.3678 (1.3455) acc 90.6250 (95.8594) lr 0.260000 +FPS@all 819.817, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:33 loss 1.3848 (1.3076) acc 96.8750 (97.3438) lr 0.260000 +epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2837 (1.3528) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 819.876, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.4823 (1.3303) acc 93.7500 (95.7812) lr 0.260000 +epoch: [80/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:10:26 loss 1.2121 (1.3533) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 819.862, TIME@all 0.312 +epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:34 loss 1.3506 (1.3001) acc 96.8750 (97.9688) lr 0.260000 +epoch: [80/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:10:27 loss 1.3188 (1.3437) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 819.778, TIME@all 0.312 +epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:06 loss 1.1982 (1.2807) acc 100.0000 (97.3438) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2367 (1.2995) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 821.662, TIME@all 0.312 +epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:07 loss 1.4024 (1.2814) acc 93.7500 (98.2812) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:10:02 loss 1.2420 (1.2847) acc 96.8750 (97.8125) lr 0.260000 +FPS@all 821.723, TIME@all 0.312 +epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2861 (1.2762) acc 93.7500 (97.3438) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:03 loss 1.2127 (1.2891) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 821.581, TIME@all 0.312 +epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2213 (1.2700) acc 100.0000 (97.9688) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:03 loss 1.3114 (1.3037) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 821.537, TIME@all 0.312 +epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2205 (1.2854) acc 100.0000 (97.9688) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.3295 (1.3011) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 821.618, TIME@all 0.312 +epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2147 (1.2601) acc 100.0000 (99.3750) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2580 (1.2909) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 821.643, TIME@all 0.312 +epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2589 (1.2899) acc 100.0000 (97.3438) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2248 (1.2886) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 821.625, TIME@all 0.312 +epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:07 loss 1.3834 (1.3068) acc 93.7500 (97.1875) lr 0.260000 +epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2640 (1.3130) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 821.611, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:09:49 loss 1.2833 (1.2481) acc 93.7500 (98.7500) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:46 loss 1.5198 (1.3024) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 821.303, TIME@all 0.312 +epoch: [82/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:09:49 loss 1.4863 (1.2737) acc 87.5000 (97.6562) lr 0.260000 +epoch: [82/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3204 (1.2986) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 821.200, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.2881 (1.2668) acc 96.8750 (97.9688) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.2417 (1.3095) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 821.113, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.4419 (1.2727) acc 93.7500 (98.2812) lr 0.260000 +epoch: [82/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3027 (1.3045) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 821.120, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.2680 (1.2725) acc 100.0000 (97.5000) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3733 (1.3107) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 821.147, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.3142 (1.2730) acc 96.8750 (97.9688) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3324 (1.3048) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.140, TIME@all 0.312 +epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.3482 (1.2613) acc 93.7500 (97.9688) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3518 (1.2861) acc 93.7500 (97.9688) lr 0.260000 +FPS@all 821.182, TIME@all 0.312 +epoch: [82/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:09:49 loss 1.2960 (1.2503) acc 100.0000 (98.9062) lr 0.260000 +epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3333 (1.2965) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 821.191, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 1:09:43 loss 1.3687 (1.2880) acc 96.8750 (97.5000) lr 0.260000 +epoch: [83/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.3875 (1.3137) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 820.042, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:09:43 loss 1.3098 (1.2762) acc 100.0000 (98.5938) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2236 (1.3127) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 819.990, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.3274 (1.2574) acc 96.8750 (98.4375) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:36 loss 1.3213 (1.2887) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 819.890, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.3876 (1.2802) acc 96.8750 (97.6562) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.3014 (1.3084) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 819.949, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:09:43 loss 1.3107 (1.3000) acc 93.7500 (97.5000) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.2524 (1.3201) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 819.926, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:09:42 loss 1.2625 (1.2624) acc 96.8750 (97.8125) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2708 (1.3086) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 820.028, TIME@all 0.312 +epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.2673 (1.2688) acc 96.8750 (97.9688) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.2945 (1.3013) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 819.991, TIME@all 0.312 +epoch: [83/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:42 loss 1.2977 (1.2840) acc 100.0000 (98.4375) lr 0.260000 +epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2728 (1.3013) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 819.985, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:31 loss 1.4490 (1.3465) acc 100.0000 (97.3438) lr 0.260000 +epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.2672 (1.3798) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 820.324, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:29 loss 1.5324 (1.3700) acc 96.8750 (96.2500) lr 0.260000 +epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.4581 (1.4079) acc 87.5000 (95.2344) lr 0.260000 +FPS@all 820.572, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:30 loss 1.6230 (1.4279) acc 90.6250 (94.5312) lr 0.260000 +epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.2552 (1.4352) acc 100.0000 (94.3750) lr 0.260000 +FPS@all 820.434, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.5198 (1.3871) acc 93.7500 (95.7812) lr 0.260000 +epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2925 (1.4278) acc 100.0000 (94.3750) lr 0.260000 +FPS@all 820.387, TIME@all 0.312 +epoch: [84/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.3933 (1.3610) acc 96.8750 (95.9375) lr 0.260000 +epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2947 (1.3928) acc 96.8750 (95.3125) lr 0.260000 +FPS@all 820.440, TIME@all 0.312 +epoch: [84/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.3530 (1.3673) acc 96.8750 (95.3125) lr 0.260000 +epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2095 (1.4072) acc 100.0000 (94.6094) lr 0.260000 +FPS@all 820.416, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:30 loss 1.4328 (1.3675) acc 93.7500 (96.0938) lr 0.260000 +epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.5120 (1.3997) acc 90.6250 (95.7031) lr 0.260000 +FPS@all 820.394, TIME@all 0.312 +epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.5651 (1.4008) acc 93.7500 (94.8438) lr 0.260000 +epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.5685 (1.4197) acc 87.5000 (94.6094) lr 0.260000 +FPS@all 820.365, TIME@all 0.312 +epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6326 (1.3889) acc 87.5000 (95.6250) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:55 loss 1.5579 (1.4250) acc 87.5000 (94.6875) lr 0.260000 +FPS@all 822.116, TIME@all 0.311 +epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.5935 (1.3863) acc 87.5000 (95.0000) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.3696 (1.3996) acc 93.7500 (94.5312) lr 0.260000 +FPS@all 822.050, TIME@all 0.311 +epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:09:15 loss 1.5326 (1.3998) acc 87.5000 (94.3750) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4394 (1.4085) acc 90.6250 (94.3750) lr 0.260000 +FPS@all 822.015, TIME@all 0.311 +epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6116 (1.3903) acc 93.7500 (95.7812) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4686 (1.4062) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 822.000, TIME@all 0.311 +epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.5496 (1.3910) acc 90.6250 (96.0938) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.3804 (1.3975) acc 96.8750 (95.6250) lr 0.260000 +FPS@all 822.042, TIME@all 0.311 +epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6175 (1.4014) acc 93.7500 (94.6875) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4397 (1.4056) acc 87.5000 (94.6875) lr 0.260000 +FPS@all 822.044, TIME@all 0.311 +epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:09:15 loss 1.4392 (1.3461) acc 96.8750 (96.8750) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4384 (1.3695) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 822.054, TIME@all 0.311 +epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.4547 (1.3645) acc 93.7500 (96.5625) lr 0.260000 +epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.5754 (1.3801) acc 87.5000 (96.0156) lr 0.260000 +FPS@all 822.018, TIME@all 0.311 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.5522 (1.3546) acc 96.8750 (96.2500) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5212 (1.3792) acc 93.7500 (95.2344) lr 0.260000 +FPS@all 821.952, TIME@all 0.311 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:34 loss 1.5045 (1.3224) acc 87.5000 (96.8750) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5095 (1.3527) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 821.987, TIME@all 0.311 +epoch: [86/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.4030 (1.3179) acc 93.7500 (97.6562) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3377 (1.3548) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 821.854, TIME@all 0.311 +epoch: [86/350][20/50] time 0.314 (0.311) data 0.001 (0.012) eta 1:08:35 loss 1.3628 (1.3040) acc 100.0000 (97.1875) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.6072 (1.3486) acc 84.3750 (95.9375) lr 0.260000 +FPS@all 821.883, TIME@all 0.311 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.6154 (1.3585) acc 87.5000 (95.6250) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5044 (1.3654) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 821.808, TIME@all 0.312 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.5558 (1.3436) acc 90.6250 (96.5625) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3877 (1.3561) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 821.903, TIME@all 0.311 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:34 loss 1.6066 (1.3036) acc 90.6250 (97.6562) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.4873 (1.3406) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 821.963, TIME@all 0.311 +epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.013) eta 1:08:35 loss 1.5968 (1.3336) acc 90.6250 (96.7188) lr 0.260000 +epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3660 (1.3595) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 821.896, TIME@all 0.311 +epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:08:45 loss 1.2947 (1.3260) acc 96.8750 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3558 (1.3445) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 820.516, TIME@all 0.312 +epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:08:45 loss 1.2934 (1.2975) acc 96.8750 (97.9688) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:08:31 loss 1.2300 (1.3400) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 820.421, TIME@all 0.312 +epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:46 loss 1.2888 (1.2969) acc 96.8750 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2073 (1.3180) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 820.397, TIME@all 0.312 +epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:08:45 loss 1.3411 (1.3156) acc 96.8750 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.4084 (1.3490) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.469, TIME@all 0.312 +epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.3065 (1.3030) acc 90.6250 (96.5625) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3257 (1.3287) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.461, TIME@all 0.312 +epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.3004 (1.3051) acc 96.8750 (97.6562) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3320 (1.3335) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 820.466, TIME@all 0.312 +epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.2377 (1.2918) acc 100.0000 (97.3438) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2638 (1.3396) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 820.422, TIME@all 0.312 +epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:46 loss 1.3157 (1.2886) acc 100.0000 (97.9688) lr 0.260000 +epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2237 (1.3167) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 820.432, TIME@all 0.312 +epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.3618 (1.3117) acc 93.7500 (97.1875) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.2816 (1.3409) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 821.264, TIME@all 0.312 +epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.3292 (1.2975) acc 96.8750 (97.5000) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.2978 (1.3436) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 821.152, TIME@all 0.312 +epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.2605 (1.3251) acc 100.0000 (96.4062) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.4561 (1.3329) acc 90.6250 (96.4844) lr 0.260000 +FPS@all 821.090, TIME@all 0.312 +epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.4107 (1.3265) acc 96.8750 (96.7188) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.5540 (1.3490) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 821.061, TIME@all 0.312 +epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.4478 (1.3381) acc 93.7500 (96.8750) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.3527 (1.3515) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 821.091, TIME@all 0.312 +epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:20 loss 1.2977 (1.2818) acc 96.8750 (97.9688) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.3046 (1.3125) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.131, TIME@all 0.312 +epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.2734 (1.2961) acc 96.8750 (97.6562) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.2902 (1.3348) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 821.115, TIME@all 0.312 +epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.3849 (1.3122) acc 93.7500 (97.0312) lr 0.260000 +epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.5208 (1.3422) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 821.128, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.3139 (1.2599) acc 100.0000 (98.5938) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.3787 (1.2714) acc 93.7500 (98.1250) lr 0.260000 +FPS@all 820.261, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2691 (1.2729) acc 100.0000 (97.9688) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:03 loss 1.3913 (1.2903) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 820.270, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.5311 (1.2842) acc 90.6250 (97.6562) lr 0.260000 +epoch: [89/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:08:04 loss 1.6006 (1.2981) acc 84.3750 (97.0312) lr 0.260000 +FPS@all 820.084, TIME@all 0.312 +epoch: [89/350][20/50] time 0.308 (0.313) data 0.000 (0.011) eta 1:08:11 loss 1.1935 (1.2680) acc 96.8750 (98.1250) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4719 (1.2916) acc 87.5000 (97.1094) lr 0.260000 +FPS@all 820.095, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.1824 (1.2609) acc 100.0000 (98.2812) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4329 (1.2853) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 820.179, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2251 (1.2487) acc 100.0000 (99.0625) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4058 (1.2786) acc 90.6250 (97.9688) lr 0.260000 +FPS@all 820.203, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2715 (1.2611) acc 93.7500 (98.9062) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4786 (1.2877) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 820.185, TIME@all 0.312 +epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2685 (1.2568) acc 96.8750 (98.2812) lr 0.260000 +epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4209 (1.2913) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 820.155, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.5439 (1.3559) acc 87.5000 (95.4688) lr 0.260000 +epoch: [90/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:07:49 loss 1.3350 (1.3545) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 819.530, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.2299 (1.3141) acc 100.0000 (97.5000) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:49 loss 1.2364 (1.3258) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 819.574, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 1:07:58 loss 1.3307 (1.3318) acc 90.6250 (96.4062) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.2651 (1.3576) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 819.439, TIME@all 0.312 +epoch: [90/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:07:59 loss 1.2100 (1.3257) acc 100.0000 (96.4062) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.1838 (1.3493) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 819.406, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.3190 (1.3040) acc 100.0000 (97.1875) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.2810 (1.3403) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 819.454, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.2129 (1.3055) acc 96.8750 (97.1875) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.3431 (1.3720) acc 96.8750 (95.2344) lr 0.260000 +FPS@all 819.441, TIME@all 0.312 +epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.4487 (1.3306) acc 96.8750 (97.1875) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.2261 (1.3624) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 819.422, TIME@all 0.312 +epoch: [90/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:07:59 loss 1.2686 (1.3372) acc 100.0000 (97.5000) lr 0.260000 +epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.2824 (1.3513) acc 100.0000 (96.8750) lr 0.260000 +FPS@all 819.457, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:07:39 loss 1.2463 (1.2869) acc 96.8750 (97.6562) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:07:34 loss 1.5287 (1.3194) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 819.914, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3845 (1.2839) acc 96.8750 (98.4375) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:33 loss 1.3609 (1.2940) acc 93.7500 (97.8125) lr 0.260000 +FPS@all 819.982, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:40 loss 1.3602 (1.2967) acc 96.8750 (97.3438) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.4910 (1.3273) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 819.833, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.5551 (1.2931) acc 90.6250 (97.3438) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.6802 (1.3166) acc 84.3750 (96.5625) lr 0.260000 +FPS@all 819.852, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3266 (1.2676) acc 93.7500 (97.8125) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.5103 (1.3038) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 819.840, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3946 (1.2882) acc 90.6250 (97.3438) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:33 loss 1.4482 (1.3071) acc 90.6250 (97.0312) lr 0.260000 +FPS@all 819.894, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3420 (1.2828) acc 96.8750 (97.9688) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.3411 (1.3006) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 819.852, TIME@all 0.312 +epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.2250 (1.2694) acc 100.0000 (97.9688) lr 0.260000 +epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.2671 (1.3059) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.897, TIME@all 0.312 +epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:07:30 loss 1.2675 (1.2959) acc 100.0000 (97.5000) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.3614 (1.3201) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 819.431, TIME@all 0.312 +epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.3495 (1.2652) acc 96.8750 (97.9688) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.3245 (1.3015) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 819.358, TIME@all 0.312 +epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.4633 (1.3065) acc 96.8750 (97.0312) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:21 loss 1.3757 (1.3337) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 819.279, TIME@all 0.312 +epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:31 loss 1.2792 (1.2922) acc 100.0000 (97.5000) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:21 loss 1.4889 (1.3449) acc 90.6250 (96.1719) lr 0.260000 +FPS@all 819.288, TIME@all 0.312 +epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:31 loss 1.3702 (1.2974) acc 96.8750 (97.5000) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.4746 (1.3393) acc 87.5000 (96.4844) lr 0.260000 +FPS@all 819.321, TIME@all 0.312 +epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.4851 (1.3167) acc 90.6250 (96.5625) lr 0.260000 +epoch: [92/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:07:20 loss 1.4684 (1.3404) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 819.304, TIME@all 0.312 +epoch: [92/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:07:30 loss 1.2580 (1.2920) acc 96.8750 (97.6562) lr 0.260000 +epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:07:20 loss 1.3388 (1.3243) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 819.344, TIME@all 0.312 +epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.3256 (1.2726) acc 96.8750 (98.1250) lr 0.260000 +epoch: [92/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:07:20 loss 1.3903 (1.3098) acc 90.6250 (97.5000) lr 0.260000 +FPS@all 819.281, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3727 (1.3414) acc 96.8750 (96.2500) lr 0.260000 +epoch: [93/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.3546 (1.3767) acc 96.8750 (95.4688) lr 0.260000 +FPS@all 821.150, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3041 (1.3461) acc 100.0000 (97.1875) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.5583 (1.3787) acc 90.6250 (95.9375) lr 0.260000 +FPS@all 821.188, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.4923 (1.3568) acc 93.7500 (96.2500) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.2709 (1.3865) acc 96.8750 (95.1562) lr 0.260000 +FPS@all 821.045, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3783 (1.3513) acc 93.7500 (96.8750) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.3711 (1.3628) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 821.100, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.4107 (1.3638) acc 93.7500 (95.3125) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.2992 (1.3619) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 821.099, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.2889 (1.3417) acc 100.0000 (96.8750) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.4701 (1.3741) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 821.041, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.3569 (1.3473) acc 93.7500 (96.7188) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.4054 (1.3564) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.090, TIME@all 0.312 +epoch: [93/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:07:04 loss 1.4200 (1.3471) acc 93.7500 (95.9375) lr 0.260000 +epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.2905 (1.3633) acc 100.0000 (96.0156) lr 0.260000 +FPS@all 821.083, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:55 loss 1.2759 (1.2900) acc 96.8750 (97.6562) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:06:42 loss 1.2461 (1.3073) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 820.671, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 1:06:56 loss 1.2898 (1.3100) acc 100.0000 (97.3438) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2879 (1.3147) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 820.498, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:06:56 loss 1.3413 (1.2822) acc 93.7500 (97.0312) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2174 (1.3105) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.501, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3123 (1.2752) acc 96.8750 (97.9688) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:06:42 loss 1.4214 (1.3147) acc 90.6250 (96.7188) lr 0.260000 +FPS@all 820.562, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 1:06:56 loss 1.2816 (1.2916) acc 96.8750 (97.5000) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2117 (1.3176) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 820.531, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3561 (1.3048) acc 90.6250 (97.3438) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:06:43 loss 1.3406 (1.3221) acc 90.6250 (96.2500) lr 0.260000 +FPS@all 820.493, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.2684 (1.2984) acc 100.0000 (97.8125) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:43 loss 1.3770 (1.3088) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 820.503, TIME@all 0.312 +epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3013 (1.3001) acc 96.8750 (97.9688) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +epoch: [94/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:06:43 loss 1.3277 (1.3074) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 820.525, TIME@all 0.312 +epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.4244 (1.2732) acc 93.7500 (98.5938) lr 0.260000 +epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3628 (1.3148) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 820.571, TIME@all 0.312 +epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2472 (1.2752) acc 100.0000 (98.2812) lr 0.260000 +epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:25 loss 1.3724 (1.3053) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 820.637, TIME@all 0.312 +epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:06:38 loss 1.2208 (1.3050) acc 100.0000 (97.3438) lr 0.260000 +epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.4278 (1.3236) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 820.477, TIME@all 0.312 +epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2183 (1.2467) acc 100.0000 (98.7500) lr 0.260000 +epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3290 (1.2837) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 820.514, TIME@all 0.312 +epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2717 (1.2882) acc 100.0000 (97.3438) lr 0.260000 +epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3624 (1.3248) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 820.497, TIME@all 0.312 +epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:39 loss 1.4073 (1.3199) acc 96.8750 (96.4062) lr 0.260000 +epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3820 (1.3267) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 820.545, TIME@all 0.312 +epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2084 (1.2715) acc 100.0000 (97.9688) lr 0.260000 +epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3962 (1.3004) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 820.566, TIME@all 0.312 +epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:06:39 loss 1.2627 (1.2680) acc 96.8750 (98.2812) lr 0.260000 +epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3227 (1.3051) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 820.531, TIME@all 0.312 +epoch: [96/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:06:21 loss 1.4948 (1.3028) acc 93.7500 (97.3438) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.3048 (1.3058) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 820.658, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:06:21 loss 1.3000 (1.3014) acc 96.8750 (97.1875) lr 0.260000 +epoch: [96/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:12 loss 1.2254 (1.2923) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 820.689, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:06:22 loss 1.4184 (1.3096) acc 96.8750 (96.8750) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:13 loss 1.2510 (1.3018) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 820.539, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:06:22 loss 1.3701 (1.2783) acc 96.8750 (97.8125) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:13 loss 1.3490 (1.3048) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 820.593, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:06:22 loss 1.2958 (1.2803) acc 100.0000 (97.3438) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.2604 (1.3267) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 820.570, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:06:21 loss 1.2972 (1.2665) acc 100.0000 (97.9688) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.2572 (1.2975) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 820.655, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:06:22 loss 1.5020 (1.3043) acc 90.6250 (97.1875) lr 0.260000 +epoch: [96/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:13 loss 1.3188 (1.3191) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 820.579, TIME@all 0.312 +epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:06:22 loss 1.3324 (1.3056) acc 96.8750 (98.1250) lr 0.260000 +epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:06:13 loss 1.1904 (1.3112) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 820.579, TIME@all 0.312 +epoch: [97/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:06:13 loss 1.2285 (1.2724) acc 100.0000 (97.6562) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.3596 (1.3012) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 820.097, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 1:06:13 loss 1.3221 (1.2738) acc 93.7500 (97.6562) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.2931 (1.2966) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 820.158, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:06:13 loss 1.1636 (1.2793) acc 100.0000 (98.4375) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.4218 (1.3095) acc 96.8750 (97.8125) lr 0.260000 +FPS@all 820.031, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:06:14 loss 1.2424 (1.2920) acc 100.0000 (97.1875) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.3469 (1.3185) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 820.019, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.012) eta 1:06:13 loss 1.2069 (1.2996) acc 100.0000 (97.1875) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.3672 (1.2982) acc 90.6250 (97.3438) lr 0.260000 +FPS@all 820.054, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 1:06:14 loss 1.2756 (1.2902) acc 93.7500 (97.8125) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:06:00 loss 1.3533 (1.3182) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.058, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:06:13 loss 1.2352 (1.2813) acc 100.0000 (97.9688) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:06:00 loss 1.3637 (1.3092) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 820.046, TIME@all 0.312 +epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:06:13 loss 1.2031 (1.2730) acc 100.0000 (97.5000) lr 0.260000 +epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.3744 (1.3250) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 820.114, TIME@all 0.312 +epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:45 loss 1.5092 (1.3662) acc 90.6250 (94.8438) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:39 loss 1.4616 (1.3971) acc 90.6250 (94.9219) lr 0.260000 +FPS@all 820.462, TIME@all 0.312 +epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.9022 (1.3556) acc 81.2500 (96.4062) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:39 loss 1.2992 (1.3807) acc 96.8750 (95.5469) lr 0.260000 +FPS@all 820.394, TIME@all 0.312 +epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.7564 (1.3665) acc 87.5000 (96.2500) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.2598 (1.3858) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 820.360, TIME@all 0.312 +epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.7568 (1.3720) acc 84.3750 (96.2500) lr 0.260000 +epoch: [98/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4500 (1.4096) acc 90.6250 (95.0781) lr 0.260000 +FPS@all 820.323, TIME@all 0.312 +epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.8032 (1.3666) acc 84.3750 (95.6250) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4621 (1.3699) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 820.373, TIME@all 0.312 +epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.6231 (1.3486) acc 93.7500 (96.2500) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4843 (1.3775) acc 96.8750 (95.7031) lr 0.260000 +FPS@all 820.369, TIME@all 0.312 +epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.6145 (1.3608) acc 87.5000 (96.5625) lr 0.260000 +epoch: [98/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4270 (1.3902) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 820.333, TIME@all 0.312 +epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.5668 (1.3669) acc 93.7500 (95.4688) lr 0.260000 +epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:40 loss 1.4603 (1.3980) acc 96.8750 (95.0781) lr 0.260000 +FPS@all 820.358, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.4851 (1.3300) acc 93.7500 (97.1875) lr 0.260000 +epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.5543 (1.3491) acc 87.5000 (96.4062) lr 0.260000 +FPS@all 820.243, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.312) data 0.000 (0.014) eta 1:05:30 loss 1.2120 (1.3059) acc 100.0000 (97.9688) lr 0.260000 +epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:27 loss 1.2284 (1.3362) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 820.296, TIME@all 0.312 +epoch: [99/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.3125 (1.3045) acc 96.8750 (97.0312) lr 0.260000 +epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.4017 (1.3444) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 820.156, TIME@all 0.312 +epoch: [99/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:05:32 loss 1.6102 (1.3425) acc 93.7500 (97.1875) lr 0.260000 +epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:05:28 loss 1.2319 (1.3507) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.139, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.4480 (1.3019) acc 96.8750 (97.3438) lr 0.260000 +epoch: [99/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.4224 (1.3348) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 820.185, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.3979 (1.3247) acc 90.6250 (96.4062) lr 0.260000 +epoch: [99/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.2784 (1.3321) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 820.207, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.2994 (1.2960) acc 96.8750 (97.6562) lr 0.260000 +epoch: [99/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.3217 (1.3244) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 820.199, TIME@all 0.312 +epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.6389 (1.3304) acc 87.5000 (97.1875) lr 0.260000 +epoch: [99/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.4749 (1.3413) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 820.168, TIME@all 0.312 +epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:05:08 loss 1.1902 (1.2903) acc 100.0000 (97.1875) lr 0.260000 +epoch: [100/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.3324 (1.3145) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 820.740, TIME@all 0.312 +epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.4519 (1.2964) acc 93.7500 (98.4375) lr 0.260000 +epoch: [100/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.4419 (1.3058) acc 90.6250 (97.7344) lr 0.260000 +FPS@all 820.764, TIME@all 0.312 +epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.5024 (1.2906) acc 90.6250 (97.6562) lr 0.260000 +epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.4901 (1.3308) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.643, TIME@all 0.312 +epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.2125 (1.2887) acc 100.0000 (98.2812) lr 0.260000 +epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2789 (1.3143) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 820.668, TIME@all 0.312 +epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.3967 (1.2757) acc 96.8750 (98.4375) lr 0.260000 +epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2354 (1.3078) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 820.690, TIME@all 0.312 +epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.3629 (1.2721) acc 93.7500 (98.2812) lr 0.260000 +epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.3837 (1.3135) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 820.697, TIME@all 0.312 +epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:08 loss 1.4032 (1.2787) acc 96.8750 (97.0312) lr 0.260000 +epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2085 (1.2976) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 820.658, TIME@all 0.312 +epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:08 loss 1.3334 (1.3248) acc 96.8750 (96.8750) lr 0.260000 +epoch: [100/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:05:04 loss 1.4364 (1.3361) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 820.698, TIME@all 0.312 +epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.3393 (1.3466) acc 93.7500 (96.5625) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:04 loss 1.3731 (1.3439) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 818.624, TIME@all 0.313 +epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.3456 (1.3022) acc 100.0000 (97.3438) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.2948 (1.3391) acc 100.0000 (96.3281) lr 0.260000 +FPS@all 818.429, TIME@all 0.313 +epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.3008 (1.3352) acc 96.8750 (97.0312) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.3647 (1.3284) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 818.444, TIME@all 0.313 +epoch: [101/350][20/50] time 0.312 (0.314) data 0.001 (0.012) eta 1:05:16 loss 1.3193 (1.3150) acc 96.8750 (95.9375) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:05:05 loss 1.2225 (1.3476) acc 100.0000 (95.7031) lr 0.260000 +FPS@all 818.503, TIME@all 0.313 +epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.2879 (1.2900) acc 93.7500 (97.3438) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.3732 (1.3330) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 818.481, TIME@all 0.313 +epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.2911 (1.2951) acc 96.8750 (97.5000) lr 0.260000 +epoch: [101/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:05:04 loss 1.2922 (1.3249) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 818.489, TIME@all 0.313 +epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.2440 (1.3124) acc 93.7500 (96.8750) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.2210 (1.3456) acc 100.0000 (95.8594) lr 0.260000 +FPS@all 818.513, TIME@all 0.313 +epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.3267 (1.3230) acc 96.8750 (96.7188) lr 0.260000 +epoch: [101/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:05:04 loss 1.2838 (1.3390) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 818.508, TIME@all 0.313 +epoch: [102/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:04:41 loss 1.2926 (1.3163) acc 96.8750 (96.7188) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:32 loss 1.3628 (1.3255) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 821.052, TIME@all 0.312 +epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.4345 (1.3241) acc 93.7500 (97.0312) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3502 (1.3376) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.958, TIME@all 0.312 +epoch: [102/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.3269 (1.3186) acc 100.0000 (96.7188) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3369 (1.3295) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.893, TIME@all 0.312 +epoch: [102/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:04:42 loss 1.3068 (1.3129) acc 93.7500 (96.7188) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3036 (1.3234) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 820.931, TIME@all 0.312 +epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.2831 (1.3192) acc 100.0000 (97.1875) lr 0.260000 +epoch: [102/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.4169 (1.3414) acc 90.6250 (96.7188) lr 0.260000 +FPS@all 820.881, TIME@all 0.312 +epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.2868 (1.3204) acc 100.0000 (97.6562) lr 0.260000 +epoch: [102/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2994 (1.3464) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 820.936, TIME@all 0.312 +epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.4019 (1.3141) acc 100.0000 (97.1875) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2717 (1.3354) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 820.920, TIME@all 0.312 +epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:41 loss 1.2799 (1.3118) acc 96.8750 (97.5000) lr 0.260000 +epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2555 (1.3354) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 820.976, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:30 loss 1.4086 (1.2671) acc 96.8750 (98.7500) lr 0.260000 +epoch: [103/350][40/50] time 0.321 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3189 (1.3052) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 819.878, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:31 loss 1.2087 (1.2627) acc 100.0000 (97.8125) lr 0.260000 +epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:04:25 loss 1.2752 (1.3008) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.884, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.2984 (1.2669) acc 100.0000 (98.9062) lr 0.260000 +epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3200 (1.2981) acc 93.7500 (97.8125) lr 0.260000 +FPS@all 819.793, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.3683 (1.2583) acc 96.8750 (98.4375) lr 0.260000 +epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.2412 (1.2857) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 819.732, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.4226 (1.2895) acc 96.8750 (97.5000) lr 0.260000 +epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3196 (1.2993) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 819.760, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.3887 (1.2705) acc 90.6250 (97.5000) lr 0.260000 +epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.2909 (1.2855) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 819.805, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:30 loss 1.2537 (1.2907) acc 96.8750 (97.9688) lr 0.260000 +epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:04:26 loss 1.2879 (1.3019) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 819.835, TIME@all 0.312 +epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:31 loss 1.2229 (1.2589) acc 100.0000 (98.2812) lr 0.260000 +epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:04:26 loss 1.1632 (1.2827) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 819.816, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.001 (0.014) eta 1:04:20 loss 1.2719 (1.2964) acc 100.0000 (98.2812) lr 0.260000 +epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.3325 (1.3187) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 820.117, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.3341 (1.2934) acc 96.8750 (97.3438) lr 0.260000 +epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.5492 (1.3262) acc 84.3750 (96.3281) lr 0.260000 +FPS@all 819.956, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.2478 (1.3276) acc 100.0000 (97.0312) lr 0.260000 +epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2313 (1.3327) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 819.998, TIME@all 0.312 +epoch: [104/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:04:20 loss 1.4380 (1.2953) acc 93.7500 (96.8750) lr 0.260000 +epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2349 (1.3197) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 820.076, TIME@all 0.312 +epoch: [104/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:04:21 loss 1.3406 (1.2760) acc 93.7500 (97.9688) lr 0.260000 +epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:04:10 loss 1.3314 (1.3114) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.964, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:04:20 loss 1.2373 (1.2757) acc 100.0000 (97.9688) lr 0.260000 +epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:04:10 loss 1.3186 (1.3062) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 820.016, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.3294 (1.2722) acc 96.8750 (97.0312) lr 0.260000 +epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.3229 (1.2967) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 819.966, TIME@all 0.312 +epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:20 loss 1.3880 (1.2933) acc 96.8750 (97.3438) lr 0.260000 +epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2493 (1.3163) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 820.013, TIME@all 0.312 +epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:51 loss 1.2672 (1.2763) acc 96.8750 (97.9688) lr 0.260000 +epoch: [105/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.2732 (1.2900) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 821.076, TIME@all 0.312 +epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:51 loss 1.3242 (1.2554) acc 96.8750 (97.6562) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.2837 (1.2940) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.141, TIME@all 0.312 +epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.4335 (1.2972) acc 96.8750 (97.5000) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.3193 (1.3119) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 821.000, TIME@all 0.312 +epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.3881 (1.2926) acc 90.6250 (97.0312) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:48 loss 1.3080 (1.2947) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 820.974, TIME@all 0.312 +epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.4120 (1.2869) acc 96.8750 (97.9688) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:48 loss 1.3797 (1.3168) acc 84.3750 (96.6406) lr 0.260000 +FPS@all 821.028, TIME@all 0.312 +epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:03:51 loss 1.1946 (1.2665) acc 100.0000 (98.2812) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.4341 (1.3031) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 820.973, TIME@all 0.312 +epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:03:52 loss 1.3665 (1.3069) acc 96.8750 (96.8750) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.2790 (1.3295) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 821.000, TIME@all 0.312 +epoch: [105/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 1:03:51 loss 1.2862 (1.2930) acc 96.8750 (97.5000) lr 0.260000 +epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.3939 (1.3216) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 821.084, TIME@all 0.312 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.001 (0.012) eta 1:04:16 loss 1.2748 (1.3129) acc 100.0000 (97.8125) lr 0.260000 +epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:07 loss 1.2851 (1.3007) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 815.525, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 1:04:16 loss 1.3774 (1.3139) acc 93.7500 (97.3438) lr 0.260000 +epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.007) eta 1:04:07 loss 1.3480 (1.2978) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 815.531, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:17 loss 1.3921 (1.2808) acc 96.8750 (98.4375) lr 0.260000 +epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.2208 (1.3073) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 815.369, TIME@all 0.314 +epoch: [106/350][20/50] time 0.313 (0.315) data 0.001 (0.012) eta 1:04:16 loss 1.2403 (1.2562) acc 100.0000 (99.2188) lr 0.260000 +epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.2939 (1.2915) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 815.474, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:17 loss 1.3334 (1.2879) acc 100.0000 (98.2812) lr 0.260000 +epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.3187 (1.3002) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 815.376, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:16 loss 1.2896 (1.2752) acc 100.0000 (98.7500) lr 0.260000 +epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.3388 (1.3120) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 815.433, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 1:04:16 loss 1.3297 (1.2963) acc 96.8750 (97.6562) lr 0.260000 +epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 1:04:08 loss 1.3065 (1.3038) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 815.446, TIME@all 0.314 +epoch: [106/350][20/50] time 0.312 (0.315) data 0.001 (0.013) eta 1:04:14 loss 1.3725 (1.3231) acc 96.8750 (96.4062) lr 0.260000 +epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 1:04:07 loss 1.4628 (1.3243) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 815.607, TIME@all 0.314 +epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:21 loss 1.2989 (1.3096) acc 96.8750 (97.1875) lr 0.260000 +epoch: [107/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3523 (1.3125) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 821.353, TIME@all 0.312 +epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.2416 (1.2991) acc 100.0000 (97.3438) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:17 loss 1.4005 (1.3343) acc 90.6250 (96.1719) lr 0.260000 +FPS@all 821.200, TIME@all 0.312 +epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.2605 (1.2976) acc 96.8750 (96.7188) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3084 (1.3076) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 821.256, TIME@all 0.312 +epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.011) eta 1:03:22 loss 1.2385 (1.2858) acc 100.0000 (97.9688) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3749 (1.3060) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 821.182, TIME@all 0.312 +epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.3015 (1.2731) acc 96.8750 (97.1875) lr 0.260000 +epoch: [107/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3953 (1.2955) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 821.249, TIME@all 0.312 +epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:03:21 loss 1.3987 (1.2931) acc 90.6250 (97.1875) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3121 (1.2947) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 821.259, TIME@all 0.312 +epoch: [107/350][20/50] time 0.309 (0.312) data 0.001 (0.012) eta 1:03:22 loss 1.3501 (1.2940) acc 96.8750 (97.8125) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.2283 (1.3140) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 821.242, TIME@all 0.312 +epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.1985 (1.2672) acc 100.0000 (97.9688) lr 0.260000 +epoch: [107/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 1:03:16 loss 1.3753 (1.2925) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 821.261, TIME@all 0.312 +epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:03:06 loss 1.2928 (1.2514) acc 100.0000 (98.2812) lr 0.260000 +epoch: [108/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3675 (1.2998) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 821.598, TIME@all 0.312 +epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:03:05 loss 1.3215 (1.2817) acc 100.0000 (98.1250) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3468 (1.3032) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 821.666, TIME@all 0.312 +epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:03:06 loss 1.2597 (1.2765) acc 96.8750 (97.6562) lr 0.260000 +epoch: [108/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:02:58 loss 1.4482 (1.3111) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 821.486, TIME@all 0.312 +epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:03:05 loss 1.3995 (1.2756) acc 93.7500 (97.8125) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3653 (1.2983) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 821.547, TIME@all 0.312 +epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:03:05 loss 1.3464 (1.2906) acc 96.8750 (97.1875) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3258 (1.3197) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 821.613, TIME@all 0.312 +epoch: [108/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:03:05 loss 1.2986 (1.2982) acc 100.0000 (97.9688) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3033 (1.3328) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 821.596, TIME@all 0.312 +epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:03:06 loss 1.2895 (1.2709) acc 96.8750 (97.9688) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3372 (1.3036) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 821.590, TIME@all 0.312 +epoch: [108/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 1:03:06 loss 1.3147 (1.2857) acc 100.0000 (97.0312) lr 0.260000 +epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3845 (1.3232) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 821.570, TIME@all 0.312 +epoch: [109/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.4079 (1.2672) acc 87.5000 (97.1875) lr 0.260000 +epoch: [109/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2820 (1.2770) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 819.372, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.2075 (1.2623) acc 100.0000 (97.5000) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2956 (1.2778) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 819.407, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:57 loss 1.2983 (1.2730) acc 100.0000 (97.9688) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.2543 (1.2759) acc 100.0000 (98.4375) lr 0.260000 +FPS@all 819.244, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 1:02:57 loss 1.2580 (1.2536) acc 100.0000 (98.1250) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:02:54 loss 1.2125 (1.2843) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 819.252, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.2619 (1.2539) acc 100.0000 (98.2812) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2680 (1.2628) acc 93.7500 (98.0469) lr 0.260000 +FPS@all 819.350, TIME@all 0.312 +epoch: [109/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:02:56 loss 1.2335 (1.2470) acc 100.0000 (98.1250) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.3225 (1.2693) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 819.297, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.3270 (1.2614) acc 96.8750 (98.1250) lr 0.260000 +epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.2867 (1.2681) acc 100.0000 (98.0469) lr 0.260000 +FPS@all 819.329, TIME@all 0.312 +epoch: [109/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 1:02:56 loss 1.2710 (1.2659) acc 96.8750 (97.6562) lr 0.260000 +epoch: [109/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.3296 (1.2674) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 819.328, TIME@all 0.312 +epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2390 (1.2771) acc 100.0000 (97.9688) lr 0.260000 +epoch: [110/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:02:25 loss 1.2898 (1.3170) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 821.918, TIME@all 0.311 +epoch: [110/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:02:35 loss 1.4564 (1.2718) acc 93.7500 (97.6562) lr 0.260000 +epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2603 (1.3107) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 821.817, TIME@all 0.312 +epoch: [110/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.1896 (1.2812) acc 100.0000 (97.8125) lr 0.260000 +epoch: [110/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2469 (1.3128) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 821.728, TIME@all 0.312 +epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:02:34 loss 1.2467 (1.2515) acc 96.8750 (98.4375) lr 0.260000 +epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:02:26 loss 1.3323 (1.2910) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 821.771, TIME@all 0.312 +epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.4556 (1.2660) acc 90.6250 (98.1250) lr 0.260000 +epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.3416 (1.3183) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 821.736, TIME@all 0.312 +epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2264 (1.2929) acc 100.0000 (97.6562) lr 0.260000 +epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2662 (1.3195) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 821.802, TIME@all 0.312 +epoch: [110/350][20/50] time 0.306 (0.312) data 0.000 (0.013) eta 1:02:34 loss 1.6545 (1.2957) acc 87.5000 (97.3438) lr 0.260000 +epoch: [110/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:02:26 loss 1.2786 (1.3213) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 821.779, TIME@all 0.312 +epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2798 (1.2516) acc 96.8750 (98.5938) lr 0.260000 +epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.3159 (1.2933) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 821.740, TIME@all 0.312 +epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:02:20 loss 1.2775 (1.2605) acc 100.0000 (98.2812) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.3773 (1.2753) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 819.336, TIME@all 0.312 +epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2542 (1.2555) acc 100.0000 (98.5938) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2964 (1.2806) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 819.318, TIME@all 0.312 +epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2380 (1.2762) acc 100.0000 (98.2812) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2824 (1.3002) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 819.312, TIME@all 0.312 +epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2951 (1.2475) acc 100.0000 (98.7500) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2150 (1.2700) acc 100.0000 (98.0469) lr 0.260000 +FPS@all 819.378, TIME@all 0.312 +epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2037 (1.2354) acc 100.0000 (98.7500) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.4882 (1.2713) acc 90.6250 (97.7344) lr 0.260000 +FPS@all 819.336, TIME@all 0.312 +epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:21 loss 1.3993 (1.2438) acc 93.7500 (98.7500) lr 0.260000 +epoch: [111/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2614 (1.2814) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 819.296, TIME@all 0.312 +epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:02:20 loss 1.1800 (1.2370) acc 100.0000 (99.0625) lr 0.260000 +epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:02:25 loss 1.2677 (1.2843) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 819.353, TIME@all 0.312 +epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2120 (1.2422) acc 100.0000 (98.9062) lr 0.260000 +epoch: [111/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.3454 (1.2790) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 819.310, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.3796 (1.2683) acc 96.8750 (97.9688) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:58 loss 1.2161 (1.3164) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 821.434, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:02:07 loss 1.1972 (1.2471) acc 100.0000 (98.7500) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.1925 (1.2759) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 821.379, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:02:07 loss 1.3615 (1.2810) acc 93.7500 (96.5625) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:01:59 loss 1.2362 (1.3029) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 821.296, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:02:08 loss 1.3708 (1.2775) acc 100.0000 (97.3438) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2891 (1.2968) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 821.244, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.3382 (1.2364) acc 100.0000 (98.5938) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2246 (1.2744) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 821.330, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.2269 (1.2665) acc 100.0000 (98.5938) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2282 (1.2804) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 821.319, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:02:08 loss 1.3315 (1.2616) acc 100.0000 (98.5938) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.1708 (1.2857) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 821.250, TIME@all 0.312 +epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.2634 (1.2681) acc 96.8750 (97.5000) lr 0.260000 +epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2294 (1.2981) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 821.326, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:01:52 loss 1.3244 (1.3045) acc 100.0000 (97.8125) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2792 (1.3121) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 820.674, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:01:52 loss 1.2143 (1.2646) acc 100.0000 (98.4375) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2731 (1.2920) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 820.589, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:52 loss 1.2775 (1.2975) acc 100.0000 (96.8750) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3135 (1.3067) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 820.544, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:53 loss 1.2692 (1.2856) acc 100.0000 (97.6562) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3105 (1.3112) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 820.457, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:52 loss 1.3065 (1.2824) acc 96.8750 (97.8125) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2250 (1.2967) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 820.556, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:01:53 loss 1.3436 (1.2942) acc 93.7500 (97.5000) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.2478 (1.3075) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 820.514, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:01:52 loss 1.3839 (1.2913) acc 90.6250 (96.8750) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.3872 (1.3134) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 820.540, TIME@all 0.312 +epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:53 loss 1.2050 (1.2838) acc 100.0000 (97.3438) lr 0.260000 +epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3117 (1.3048) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 820.532, TIME@all 0.312 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:01:52 loss 1.4046 (1.2909) acc 90.6250 (97.6562) lr 0.260000 +epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 1:01:43 loss 1.3622 (1.3257) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 818.185, TIME@all 0.313 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3800 (1.2879) acc 90.6250 (97.5000) lr 0.260000 +epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.3842 (1.3142) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 818.025, TIME@all 0.313 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3237 (1.2854) acc 93.7500 (97.9688) lr 0.260000 +epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.2472 (1.2967) acc 100.0000 (97.8125) lr 0.260000 +FPS@all 817.990, TIME@all 0.313 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.4853 (1.2928) acc 93.7500 (97.5000) lr 0.260000 +epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3297 (1.2980) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 818.080, TIME@all 0.313 +epoch: [114/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3126 (1.2998) acc 96.8750 (98.2812) lr 0.260000 +epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3668 (1.3136) acc 96.8750 (97.8906) lr 0.260000 +FPS@all 818.069, TIME@all 0.313 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:01:52 loss 1.4265 (1.2916) acc 96.8750 (97.6562) lr 0.260000 +epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.2445 (1.3211) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 818.061, TIME@all 0.313 +epoch: [114/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3725 (1.2887) acc 96.8750 (97.1875) lr 0.260000 +epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.4530 (1.3271) acc 93.7500 (96.4844) lr 0.260000 +FPS@all 818.068, TIME@all 0.313 +epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.5391 (1.3278) acc 87.5000 (96.4062) lr 0.260000 +epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3430 (1.3268) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 818.098, TIME@all 0.313 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.2804 (1.2959) acc 100.0000 (97.5000) lr 0.260000 +epoch: [115/350][40/50] time 0.314 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.2491 (1.3378) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 814.499, TIME@all 0.314 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.3182 (1.2841) acc 100.0000 (98.1250) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.007) eta 1:01:39 loss 1.4364 (1.3273) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 814.551, TIME@all 0.314 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.4814 (1.3285) acc 84.3750 (96.4062) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.4380 (1.3397) acc 93.7500 (96.1719) lr 0.260000 +FPS@all 814.397, TIME@all 0.314 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:43 loss 1.4958 (1.3269) acc 93.7500 (96.7188) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.4416 (1.3369) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 814.404, TIME@all 0.314 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:43 loss 1.4680 (1.3086) acc 96.8750 (97.1875) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.001 (0.007) eta 1:01:40 loss 1.3314 (1.3367) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 814.434, TIME@all 0.314 +epoch: [115/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:01:42 loss 1.4629 (1.3321) acc 90.6250 (96.5625) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.3963 (1.3730) acc 96.8750 (94.8438) lr 0.260000 +FPS@all 814.504, TIME@all 0.314 +epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:43 loss 1.3783 (1.2959) acc 96.8750 (97.5000) lr 0.260000 +epoch: [115/350][40/50] time 0.316 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.3888 (1.3176) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 814.413, TIME@all 0.314 +epoch: [115/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:01:42 loss 1.3629 (1.2881) acc 96.8750 (97.9688) lr 0.260000 +epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.3878 (1.3269) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 814.450, TIME@all 0.314 +epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.2708 (1.3306) acc 100.0000 (97.3438) lr 0.260000 +epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:58 loss 1.3687 (1.3518) acc 100.0000 (96.2500) lr 0.260000 +FPS@all 808.942, TIME@all 0.316 +epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.013) eta 1:01:55 loss 1.2462 (1.3113) acc 100.0000 (96.7188) lr 0.260000 +epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.007) eta 1:01:58 loss 1.3786 (1.3251) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 808.996, TIME@all 0.316 +epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3164 (1.3163) acc 96.8750 (96.8750) lr 0.260000 +epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.006) eta 1:01:58 loss 1.4332 (1.3426) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 808.918, TIME@all 0.316 +epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.012) eta 1:01:56 loss 1.2616 (1.2947) acc 96.8750 (97.0312) lr 0.260000 +epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.4820 (1.3417) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 808.857, TIME@all 0.316 +epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.013) eta 1:01:56 loss 1.3617 (1.3070) acc 96.8750 (96.7188) lr 0.260000 +epoch: [116/350][40/50] time 0.321 (0.318) data 0.001 (0.007) eta 1:01:59 loss 1.3308 (1.3298) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 808.903, TIME@all 0.316 +epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.2199 (1.3005) acc 100.0000 (97.5000) lr 0.260000 +epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.6124 (1.3370) acc 90.6250 (96.4062) lr 0.260000 +FPS@all 808.935, TIME@all 0.316 +epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3787 (1.2980) acc 96.8750 (97.3438) lr 0.260000 +epoch: [116/350][40/50] time 0.321 (0.318) data 0.001 (0.006) eta 1:01:58 loss 1.2107 (1.3307) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 809.061, TIME@all 0.316 +epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3660 (1.3456) acc 96.8750 (97.0312) lr 0.260000 +epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.2776 (1.3398) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 808.928, TIME@all 0.316 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.4747 (1.3426) acc 93.7500 (97.0312) lr 0.260000 +epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:49 loss 1.3984 (1.3744) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 820.078, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:00:52 loss 1.3679 (1.3169) acc 93.7500 (97.0312) lr 0.260000 +epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:50 loss 1.3606 (1.3570) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 819.999, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.6747 (1.3392) acc 84.3750 (96.8750) lr 0.260000 +epoch: [117/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.3275 (1.3514) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 820.035, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5852 (1.3146) acc 90.6250 (97.1875) lr 0.260000 +epoch: [117/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.2231 (1.3554) acc 100.0000 (95.7812) lr 0.260000 +FPS@all 819.987, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5490 (1.3395) acc 93.7500 (96.0938) lr 0.260000 +epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.3245 (1.3590) acc 100.0000 (95.2344) lr 0.260000 +FPS@all 820.040, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5028 (1.3347) acc 93.7500 (97.6562) lr 0.260000 +epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.4873 (1.3742) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 820.033, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.4668 (1.3120) acc 93.7500 (97.0312) lr 0.260000 +epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.2063 (1.3408) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 820.061, TIME@all 0.312 +epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:00:52 loss 1.4449 (1.3136) acc 93.7500 (97.3438) lr 0.260000 +epoch: [117/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:00:50 loss 1.2417 (1.3361) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 820.032, TIME@all 0.312 +epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.2721 (1.3580) acc 100.0000 (95.6250) lr 0.260000 +epoch: [118/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3493 (1.3868) acc 93.7500 (95.0781) lr 0.260000 +FPS@all 819.375, TIME@all 0.312 +epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.011) eta 1:00:46 loss 1.4088 (1.3676) acc 93.7500 (95.9375) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.2988 (1.3649) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 819.250, TIME@all 0.312 +epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3231 (1.3699) acc 93.7500 (95.6250) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3168 (1.3950) acc 100.0000 (95.6250) lr 0.260000 +FPS@all 819.318, TIME@all 0.312 +epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3379 (1.3293) acc 96.8750 (97.9688) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3785 (1.3671) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 819.215, TIME@all 0.312 +epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3007 (1.3496) acc 96.8750 (96.2500) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3858 (1.3662) acc 96.8750 (95.7812) lr 0.260000 +FPS@all 819.299, TIME@all 0.312 +epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:47 loss 1.5250 (1.3488) acc 93.7500 (96.7188) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:00:37 loss 1.3251 (1.3813) acc 100.0000 (95.4688) lr 0.260000 +FPS@all 819.252, TIME@all 0.312 +epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:47 loss 1.3959 (1.3178) acc 96.8750 (96.7188) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.4385 (1.3757) acc 93.7500 (94.8438) lr 0.260000 +FPS@all 819.240, TIME@all 0.312 +epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 1:00:46 loss 1.2193 (1.3481) acc 100.0000 (96.7188) lr 0.260000 +epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:37 loss 1.4089 (1.3719) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 819.266, TIME@all 0.312 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3287 (1.2880) acc 100.0000 (97.3438) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3692 (1.3123) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 819.292, TIME@all 0.312 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.012) eta 1:00:25 loss 1.2496 (1.3192) acc 96.8750 (96.5625) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:00:20 loss 1.2500 (1.3276) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 819.170, TIME@all 0.313 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3323 (1.2889) acc 93.7500 (97.3438) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.2911 (1.3215) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.173, TIME@all 0.313 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.2662 (1.2880) acc 100.0000 (97.5000) lr 0.260000 +epoch: [119/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3449 (1.3266) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 819.240, TIME@all 0.312 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.1858 (1.2819) acc 100.0000 (97.1875) lr 0.260000 +epoch: [119/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3603 (1.3055) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 819.188, TIME@all 0.313 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.2217 (1.3051) acc 100.0000 (97.6562) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.1926 (1.3307) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 819.196, TIME@all 0.313 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3093 (1.3136) acc 100.0000 (96.7188) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3814 (1.3332) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 819.161, TIME@all 0.313 +epoch: [119/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 1:00:25 loss 1.2437 (1.2838) acc 100.0000 (98.1250) lr 0.260000 +epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3340 (1.3137) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 819.166, TIME@all 0.313 +epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.2227 (1.2840) acc 100.0000 (97.5000) lr 0.260000 +epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:59:49 loss 1.4357 (1.2975) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 822.018, TIME@all 0.311 +epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 1:00:05 loss 1.3144 (1.2915) acc 96.8750 (98.1250) lr 0.260000 +epoch: [120/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:59:48 loss 1.3570 (1.3045) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 822.116, TIME@all 0.311 +epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.2935 (1.2916) acc 96.8750 (97.1875) lr 0.260000 +epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2932 (1.2945) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 821.939, TIME@all 0.311 +epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.2632 (1.2841) acc 96.8750 (97.9688) lr 0.260000 +epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:50 loss 1.3486 (1.3051) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.938, TIME@all 0.311 +epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.2990 (1.2686) acc 100.0000 (98.1250) lr 0.260000 +epoch: [120/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:59:49 loss 1.3681 (1.2904) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 821.992, TIME@all 0.311 +epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:05 loss 1.2574 (1.2848) acc 96.8750 (97.8125) lr 0.260000 +epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2902 (1.2953) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 822.017, TIME@all 0.311 +epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.3383 (1.2667) acc 93.7500 (98.1250) lr 0.260000 +epoch: [120/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:59:49 loss 1.3439 (1.2980) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 821.961, TIME@all 0.311 +epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.4026 (1.2822) acc 96.8750 (97.0312) lr 0.260000 +epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2744 (1.2920) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.979, TIME@all 0.311 +epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2943 (1.2913) acc 100.0000 (97.3438) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2831 (1.3162) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 819.354, TIME@all 0.312 +epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.014) eta 0:59:52 loss 1.3365 (1.2836) acc 96.8750 (97.3438) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:59:45 loss 1.3285 (1.3047) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 819.379, TIME@all 0.312 +epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2240 (1.2517) acc 96.8750 (97.9688) lr 0.260000 +epoch: [121/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2618 (1.2798) acc 100.0000 (97.5781) lr 0.260000 +FPS@all 819.239, TIME@all 0.312 +epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3446 (1.2604) acc 96.8750 (97.8125) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.3106 (1.2866) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 819.287, TIME@all 0.312 +epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3294 (1.2741) acc 96.8750 (97.6562) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:59:46 loss 1.4999 (1.2951) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 819.222, TIME@all 0.312 +epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2792 (1.2616) acc 96.8750 (97.6562) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:59:46 loss 1.3357 (1.3007) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 819.331, TIME@all 0.312 +epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3560 (1.2941) acc 100.0000 (97.8125) lr 0.260000 +epoch: [121/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2422 (1.3239) acc 100.0000 (96.5625) lr 0.260000 +FPS@all 819.285, TIME@all 0.312 +epoch: [121/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 0:59:52 loss 1.2416 (1.2523) acc 96.8750 (98.1250) lr 0.260000 +epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2670 (1.2792) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 819.318, TIME@all 0.312 +epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:59:48 loss 1.2832 (1.2386) acc 96.8750 (98.7500) lr 0.260000 +epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:59:33 loss 1.5490 (1.2933) acc 90.6250 (97.1875) lr 0.260000 +FPS@all 819.625, TIME@all 0.312 +epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:59:48 loss 1.2892 (1.2539) acc 100.0000 (98.4375) lr 0.260000 +epoch: [122/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:32 loss 1.3166 (1.2846) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 819.680, TIME@all 0.312 +epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.3301 (1.2542) acc 93.7500 (97.6562) lr 0.260000 +epoch: [122/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4414 (1.2954) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.484, TIME@all 0.312 +epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:48 loss 1.2465 (1.2565) acc 100.0000 (98.5938) lr 0.260000 +epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4383 (1.2949) acc 93.7500 (97.5000) lr 0.260000 +FPS@all 819.541, TIME@all 0.312 +epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:48 loss 1.3643 (1.2539) acc 96.8750 (98.4375) lr 0.260000 +epoch: [122/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.3212 (1.2803) acc 96.8750 (97.7344) lr 0.260000 +FPS@all 819.560, TIME@all 0.312 +epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:59:49 loss 1.3249 (1.2661) acc 93.7500 (97.9688) lr 0.260000 +epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.3736 (1.2939) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 819.537, TIME@all 0.312 +epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.2685 (1.2629) acc 100.0000 (98.2812) lr 0.260000 +epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.2875 (1.2893) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 819.590, TIME@all 0.312 +epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.2169 (1.2429) acc 100.0000 (98.1250) lr 0.260000 +epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4188 (1.2671) acc 96.8750 (98.0469) lr 0.260000 +FPS@all 819.563, TIME@all 0.312 +epoch: [123/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.1763 (1.2905) acc 100.0000 (97.1875) lr 0.260000 +epoch: [123/350][40/50] time 0.318 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3944 (1.3205) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 813.685, TIME@all 0.315 +epoch: [123/350][20/50] time 0.311 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.3329 (1.3120) acc 96.8750 (96.7188) lr 0.260000 +epoch: [123/350][40/50] time 0.318 (0.316) data 0.001 (0.007) eta 0:59:45 loss 1.4059 (1.3452) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 813.714, TIME@all 0.315 +epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.2048 (1.3026) acc 100.0000 (97.0312) lr 0.260000 +epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.007) eta 0:59:45 loss 1.4569 (1.3276) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 813.655, TIME@all 0.315 +epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.1936 (1.3006) acc 100.0000 (97.1875) lr 0.260000 +epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3491 (1.3301) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 813.561, TIME@all 0.315 +epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:46 loss 1.2879 (1.3199) acc 100.0000 (96.5625) lr 0.260000 +epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:44 loss 1.4020 (1.3289) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 813.762, TIME@all 0.315 +epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:59:48 loss 1.3763 (1.3129) acc 96.8750 (96.8750) lr 0.260000 +epoch: [123/350][40/50] time 0.318 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3451 (1.3347) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 813.579, TIME@all 0.315 +epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.2538 (1.2984) acc 100.0000 (97.6562) lr 0.260000 +epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.4514 (1.3290) acc 93.7500 (96.8750) lr 0.260000 +FPS@all 813.620, TIME@all 0.315 +epoch: [123/350][20/50] time 0.313 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.3274 (1.2936) acc 96.8750 (97.3438) lr 0.260000 +epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.5061 (1.3250) acc 87.5000 (96.3281) lr 0.260000 +FPS@all 813.625, TIME@all 0.315 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:00 loss 1.6863 (1.3255) acc 84.3750 (96.2500) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3776 (1.3461) acc 96.8750 (95.8594) lr 0.260000 +FPS@all 820.139, TIME@all 0.312 +epoch: [124/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:59:00 loss 1.2946 (1.2852) acc 93.7500 (98.2812) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3675 (1.3167) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 820.195, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.2896 (1.2927) acc 96.8750 (97.8125) lr 0.260000 +epoch: [124/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.2710 (1.3084) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 820.021, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3264 (1.2989) acc 96.8750 (98.4375) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.2523 (1.3132) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 820.088, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.4306 (1.3033) acc 90.6250 (97.3438) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.4281 (1.3061) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 820.018, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3743 (1.2949) acc 96.8750 (97.0312) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3207 (1.3229) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 820.090, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.4536 (1.3073) acc 93.7500 (97.0312) lr 0.260000 +epoch: [124/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.2681 (1.3346) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 820.060, TIME@all 0.312 +epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3449 (1.3000) acc 96.8750 (97.1875) lr 0.260000 +epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3528 (1.3249) acc 96.8750 (96.7188) lr 0.260000 +FPS@all 820.082, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:44 loss 1.4814 (1.3093) acc 93.7500 (97.3438) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.3673 (1.3261) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 820.561, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.3864 (1.3240) acc 96.8750 (96.8750) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.3576 (1.3376) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 820.441, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:44 loss 1.3604 (1.3092) acc 93.7500 (98.2812) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.3359 (1.3231) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 820.612, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.4583 (1.3077) acc 96.8750 (97.3438) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2696 (1.3332) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 820.424, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:58:45 loss 1.5431 (1.3267) acc 93.7500 (97.0312) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.001 (0.007) eta 0:58:38 loss 1.3938 (1.3258) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 820.505, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:58:45 loss 1.5118 (1.3092) acc 96.8750 (97.8125) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.4696 (1.3400) acc 87.5000 (96.6406) lr 0.260000 +FPS@all 820.523, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.4297 (1.3231) acc 96.8750 (96.5625) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2805 (1.3232) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 820.510, TIME@all 0.312 +epoch: [125/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:58:45 loss 1.2553 (1.3056) acc 100.0000 (96.8750) lr 0.260000 +epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2579 (1.3299) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.490, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.3136 (1.2970) acc 96.8750 (97.8125) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.4177 (1.3058) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 820.366, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:58:26 loss 1.2778 (1.2775) acc 100.0000 (97.9688) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.2453 (1.2821) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 820.435, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:28 loss 1.3298 (1.3002) acc 93.7500 (97.8125) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:23 loss 1.3278 (1.3057) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 820.255, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.2207 (1.2841) acc 96.8750 (97.6562) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.3536 (1.2981) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 820.314, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:28 loss 1.1982 (1.2662) acc 100.0000 (97.9688) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.4208 (1.2860) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 820.281, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:27 loss 1.3801 (1.2865) acc 93.7500 (97.6562) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:58:23 loss 1.4050 (1.3061) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 820.247, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.3163 (1.2882) acc 100.0000 (96.8750) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.2808 (1.3070) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 820.311, TIME@all 0.312 +epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.2743 (1.2930) acc 100.0000 (97.6562) lr 0.260000 +epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.3121 (1.2992) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 820.311, TIME@all 0.312 +epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.3856 (1.2713) acc 96.8750 (97.5000) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2474 (1.3012) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 819.990, TIME@all 0.312 +epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:06 loss 1.4688 (1.2812) acc 93.7500 (97.5000) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:58:06 loss 1.2432 (1.2873) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 820.027, TIME@all 0.312 +epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.3163 (1.2685) acc 96.8750 (98.2812) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.4103 (1.3012) acc 93.7500 (97.6562) lr 0.260000 +FPS@all 819.859, TIME@all 0.312 +epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.2864 (1.2756) acc 100.0000 (97.9688) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.2566 (1.2865) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 819.881, TIME@all 0.312 +epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.4809 (1.2827) acc 96.8750 (97.9688) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2263 (1.3057) acc 100.0000 (97.3438) lr 0.260000 +FPS@all 819.965, TIME@all 0.312 +epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.3705 (1.2964) acc 96.8750 (97.6562) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.001 (0.006) eta 0:58:07 loss 1.2245 (1.3093) acc 100.0000 (97.0312) lr 0.260000 +FPS@all 819.918, TIME@all 0.312 +epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.4127 (1.2741) acc 93.7500 (97.1875) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2143 (1.2968) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 819.939, TIME@all 0.312 +epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.2780 (1.2625) acc 96.8750 (98.2812) lr 0.260000 +epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.3674 (1.2962) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 819.928, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.3176 (1.2681) acc 96.8750 (97.9688) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:57:53 loss 1.4592 (1.3228) acc 90.6250 (96.7188) lr 0.260000 +FPS@all 820.873, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.1936 (1.2852) acc 100.0000 (97.6562) lr 0.260000 +epoch: [128/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:57:52 loss 1.6532 (1.3456) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 820.924, TIME@all 0.312 +epoch: [128/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.2375 (1.3140) acc 96.8750 (96.7188) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.3706 (1.3632) acc 93.7500 (95.3125) lr 0.260000 +FPS@all 820.750, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.3144 (1.3269) acc 100.0000 (97.1875) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.4988 (1.3650) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 820.740, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.5287 (1.3154) acc 93.7500 (96.8750) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.2311 (1.3451) acc 100.0000 (96.1719) lr 0.260000 +FPS@all 820.837, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.3183 (1.2944) acc 93.7500 (97.0312) lr 0.260000 +epoch: [128/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.5662 (1.3579) acc 93.7500 (95.3906) lr 0.260000 +FPS@all 820.800, TIME@all 0.312 +epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.2441 (1.2960) acc 96.8750 (97.3438) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.6918 (1.3556) acc 84.3750 (95.9375) lr 0.260000 +FPS@all 820.813, TIME@all 0.312 +epoch: [128/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:57:58 loss 1.3340 (1.2959) acc 96.8750 (97.1875) lr 0.260000 +epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:57:53 loss 1.6742 (1.3458) acc 87.5000 (95.9375) lr 0.260000 +FPS@all 820.774, TIME@all 0.312 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:49 loss 1.2866 (1.3250) acc 100.0000 (97.5000) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4013 (1.3451) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 819.287, TIME@all 0.312 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.2991 (1.3258) acc 96.8750 (97.1875) lr 0.260000 +epoch: [129/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.3933 (1.3327) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 819.142, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.2922 (1.3251) acc 96.8750 (96.8750) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5152 (1.3377) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 819.160, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:49 loss 1.4328 (1.3236) acc 93.7500 (97.1875) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5942 (1.3470) acc 87.5000 (96.6406) lr 0.260000 +FPS@all 819.138, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:49 loss 1.2317 (1.3216) acc 96.8750 (97.1875) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5366 (1.3326) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 819.181, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.3030 (1.3364) acc 100.0000 (96.8750) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5083 (1.3616) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 819.172, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:49 loss 1.3842 (1.3166) acc 93.7500 (97.0312) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4849 (1.3453) acc 90.6250 (95.7812) lr 0.260000 +FPS@all 819.187, TIME@all 0.313 +epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:50 loss 1.3290 (1.3424) acc 96.8750 (96.4062) lr 0.260000 +epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4364 (1.3539) acc 93.7500 (95.7031) lr 0.260000 +FPS@all 819.185, TIME@all 0.313 +epoch: [130/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.4526 (1.3210) acc 93.7500 (96.7188) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:57:30 loss 1.3440 (1.3078) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 818.634, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:32 loss 1.2557 (1.3202) acc 96.8750 (95.7812) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:57:29 loss 1.3596 (1.3267) acc 96.8750 (96.1719) lr 0.260000 +FPS@all 818.713, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:33 loss 1.3131 (1.2905) acc 96.8750 (97.3438) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.3256 (1.2965) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 818.587, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.2130 (1.2631) acc 100.0000 (98.4375) lr 0.260000 +epoch: [130/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:57:30 loss 1.3250 (1.2945) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 818.602, TIME@all 0.313 +epoch: [130/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:57:33 loss 1.2682 (1.2927) acc 100.0000 (98.1250) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.2304 (1.3140) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 818.600, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:32 loss 1.1960 (1.2876) acc 100.0000 (97.5000) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:57:30 loss 1.3639 (1.2845) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 818.639, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:32 loss 1.3350 (1.2798) acc 96.8750 (97.3438) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.2937 (1.2901) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 818.562, TIME@all 0.313 +epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.3158 (1.2839) acc 93.7500 (96.8750) lr 0.260000 +epoch: [130/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:57:30 loss 1.3613 (1.2873) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 818.618, TIME@all 0.313 +epoch: [131/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:57:10 loss 1.2940 (1.3069) acc 96.8750 (96.8750) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:57:04 loss 1.2532 (1.3430) acc 96.8750 (96.2500) lr 0.260000 +FPS@all 820.678, TIME@all 0.312 +epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:12 loss 1.2310 (1.3037) acc 96.8750 (97.3438) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2454 (1.3288) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 820.401, TIME@all 0.312 +epoch: [131/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4684 (1.3167) acc 90.6250 (96.8750) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.2651 (1.3410) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 820.495, TIME@all 0.312 +epoch: [131/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.2819 (1.3243) acc 96.8750 (96.0938) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.3974 (1.3426) acc 93.7500 (96.0938) lr 0.260000 +FPS@all 820.451, TIME@all 0.312 +epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:12 loss 1.4192 (1.3218) acc 93.7500 (96.4062) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.4442 (1.3289) acc 93.7500 (96.4062) lr 0.260000 +FPS@all 820.420, TIME@all 0.312 +epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.3217 (1.3047) acc 96.8750 (97.5000) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2610 (1.3273) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.490, TIME@all 0.312 +epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4061 (1.3035) acc 93.7500 (96.4062) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2734 (1.3396) acc 100.0000 (96.0938) lr 0.260000 +FPS@all 820.458, TIME@all 0.312 +epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4465 (1.2992) acc 93.7500 (97.5000) lr 0.260000 +epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.3393 (1.3253) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 820.509, TIME@all 0.312 +epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:57:08 loss 1.6798 (1.3224) acc 90.6250 (96.8750) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.3637 (1.3538) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 817.533, TIME@all 0.313 +epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:57:09 loss 1.5463 (1.3126) acc 87.5000 (97.5000) lr 0.260000 +epoch: [132/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:57:02 loss 1.2848 (1.3330) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 817.363, TIME@all 0.313 +epoch: [132/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.3603 (1.3091) acc 96.8750 (97.0312) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2990 (1.3238) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 817.413, TIME@all 0.313 +epoch: [132/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.4834 (1.3185) acc 90.6250 (97.0312) lr 0.260000 +epoch: [132/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:57:02 loss 1.2193 (1.3454) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 817.353, TIME@all 0.313 +epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.4365 (1.3198) acc 90.6250 (96.8750) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:02 loss 1.4076 (1.3539) acc 90.6250 (95.8594) lr 0.260000 +FPS@all 817.372, TIME@all 0.313 +epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:57:08 loss 1.4646 (1.3228) acc 93.7500 (96.0938) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2791 (1.3402) acc 96.8750 (96.0938) lr 0.260000 +FPS@all 817.417, TIME@all 0.313 +epoch: [132/350][20/50] time 0.316 (0.314) data 0.001 (0.013) eta 0:57:08 loss 1.3006 (1.3230) acc 100.0000 (96.7188) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.001 (0.007) eta 0:57:01 loss 1.3082 (1.3548) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 817.387, TIME@all 0.313 +epoch: [132/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:57:08 loss 1.6315 (1.3143) acc 87.5000 (96.4062) lr 0.260000 +epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2599 (1.3417) acc 96.8750 (95.9375) lr 0.260000 +FPS@all 817.426, TIME@all 0.313 +epoch: [133/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.2503 (1.2672) acc 100.0000 (97.9688) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.4011 (1.3117) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 822.754, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.014) eta 0:56:40 loss 1.2640 (1.2802) acc 100.0000 (97.8125) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.2516 (1.2980) acc 100.0000 (97.2656) lr 0.260000 +FPS@all 822.844, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.3401 (1.3027) acc 90.6250 (97.5000) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:27 loss 1.4148 (1.3093) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 822.707, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.5672 (1.2837) acc 90.6250 (97.1875) lr 0.260000 +epoch: [133/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:56:27 loss 1.3829 (1.3078) acc 90.6250 (96.8750) lr 0.260000 +FPS@all 822.730, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:56:41 loss 1.3584 (1.2846) acc 96.8750 (98.1250) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:27 loss 1.2818 (1.3122) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 822.758, TIME@all 0.311 +epoch: [133/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.4232 (1.2703) acc 93.7500 (98.1250) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.3546 (1.2997) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 822.731, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.014) eta 0:56:41 loss 1.3347 (1.2721) acc 100.0000 (98.1250) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.3018 (1.2908) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 822.768, TIME@all 0.311 +epoch: [133/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:56:41 loss 1.3286 (1.2507) acc 96.8750 (98.5938) lr 0.260000 +epoch: [133/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:56:26 loss 1.4027 (1.2871) acc 93.7500 (97.8906) lr 0.260000 +FPS@all 822.776, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4061 (1.2348) acc 93.7500 (99.0625) lr 0.260000 +epoch: [134/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2425 (1.2460) acc 100.0000 (98.9844) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.906, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:21 loss 1.1831 (1.2354) acc 100.0000 (98.1250) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1647 (1.2502) acc 100.0000 (98.0469) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.969, TIME@all 0.311 +epoch: [134/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:56:22 loss 1.3680 (1.2684) acc 96.8750 (97.5000) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2452 (1.2740) acc 90.6250 (97.1094) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.838, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.2789 (1.2306) acc 100.0000 (98.7500) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:17 loss 1.3267 (1.2643) acc 96.8750 (97.8125) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.781, TIME@all 0.312 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4153 (1.2703) acc 87.5000 (97.1875) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2740 (1.2935) acc 96.8750 (96.7969) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.849, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.3039 (1.2297) acc 93.7500 (99.0625) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2452 (1.2470) acc 100.0000 (98.5938) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.862, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.2779 (1.2449) acc 96.8750 (98.2812) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1626 (1.2651) acc 100.0000 (97.5781) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.869, TIME@all 0.311 +epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4239 (1.2407) acc 90.6250 (98.5938) lr 0.260000 +epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1717 (1.2779) acc 100.0000 (97.1094) lr 0.260000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 821.856, TIME@all 0.311 +epoch: [135/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:56:26 loss 1.4440 (1.2633) acc 93.7500 (97.9688) lr 0.260000 +epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.2722 (1.2761) acc 96.8750 (98.1250) lr 0.260000 +FPS@all 819.169, TIME@all 0.313 +epoch: [135/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:56:26 loss 1.2236 (1.2330) acc 100.0000 (97.9688) lr 0.260000 +epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2205 (1.2618) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 819.258, TIME@all 0.312 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:56:27 loss 1.3356 (1.2560) acc 96.8750 (98.7500) lr 0.260000 +epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:11 loss 1.3273 (1.2930) acc 93.7500 (97.8125) lr 0.260000 +FPS@all 819.105, TIME@all 0.313 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:56:27 loss 1.4384 (1.2674) acc 93.7500 (97.6562) lr 0.260000 +epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.2826 (1.2810) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 819.087, TIME@all 0.313 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:56:27 loss 1.3623 (1.2796) acc 100.0000 (97.9688) lr 0.260000 +epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.5304 (1.2875) acc 90.6250 (97.8125) lr 0.260000 +FPS@all 819.110, TIME@all 0.313 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:56:27 loss 1.3413 (1.2612) acc 96.8750 (98.4375) lr 0.260000 +epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2375 (1.2651) acc 100.0000 (98.4375) lr 0.260000 +FPS@all 819.118, TIME@all 0.313 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:56:27 loss 1.3704 (1.2654) acc 93.7500 (97.6562) lr 0.260000 +epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2192 (1.2803) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 819.077, TIME@all 0.313 +epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:56:26 loss 1.2483 (1.2560) acc 100.0000 (97.6562) lr 0.260000 +epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.1857 (1.2797) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 819.163, TIME@all 0.313 +epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.3796 (1.2795) acc 100.0000 (97.6562) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2228 (1.2892) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 820.953, TIME@all 0.312 +epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:55:47 loss 1.3847 (1.2934) acc 93.7500 (97.0312) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:55:40 loss 1.3094 (1.3011) acc 93.7500 (96.5625) lr 0.260000 +FPS@all 820.838, TIME@all 0.312 +epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:55:47 loss 1.4261 (1.2493) acc 90.6250 (97.8125) lr 0.260000 +epoch: [136/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:55:41 loss 1.2702 (1.2900) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 820.821, TIME@all 0.312 +epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:48 loss 1.2572 (1.2831) acc 93.7500 (97.3438) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2319 (1.3100) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 820.854, TIME@all 0.312 +epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.3041 (1.2769) acc 96.8750 (97.9688) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2741 (1.2914) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 820.855, TIME@all 0.312 +epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2717 (1.2795) acc 100.0000 (97.5000) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2664 (1.3130) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 820.890, TIME@all 0.312 +epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2139 (1.2762) acc 100.0000 (97.3438) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2881 (1.3053) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 820.909, TIME@all 0.312 +epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2553 (1.2722) acc 96.8750 (98.4375) lr 0.260000 +epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.1817 (1.2853) acc 100.0000 (97.7344) lr 0.260000 +FPS@all 820.862, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.2248 (1.2564) acc 100.0000 (99.0625) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.1623 (1.2853) acc 100.0000 (98.1250) lr 0.260000 +FPS@all 821.770, TIME@all 0.312 +epoch: [137/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:32 loss 1.2972 (1.2748) acc 96.8750 (98.1250) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 0:55:25 loss 1.3615 (1.3033) acc 90.6250 (97.6562) lr 0.260000 +FPS@all 821.852, TIME@all 0.311 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3347 (1.2436) acc 96.8750 (99.0625) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.4410 (1.2931) acc 87.5000 (97.3438) lr 0.260000 +FPS@all 821.678, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3103 (1.2697) acc 96.8750 (97.5000) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2866 (1.3036) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 821.670, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3784 (1.2731) acc 93.7500 (98.2812) lr 0.260000 +epoch: [137/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.3210 (1.3030) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 821.678, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:32 loss 1.4110 (1.2848) acc 93.7500 (97.3438) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2853 (1.3098) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.741, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.1759 (1.2557) acc 100.0000 (98.7500) lr 0.260000 +epoch: [137/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2404 (1.2819) acc 100.0000 (98.0469) lr 0.260000 +FPS@all 821.711, TIME@all 0.312 +epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.4330 (1.2849) acc 93.7500 (97.8125) lr 0.260000 +epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2648 (1.3102) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 821.724, TIME@all 0.312 +epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:55:23 loss 1.2661 (1.2619) acc 96.8750 (98.2812) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3513 (1.3061) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 821.387, TIME@all 0.312 +epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2147 (1.2722) acc 100.0000 (97.5000) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3693 (1.3188) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 821.433, TIME@all 0.312 +epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2965 (1.2864) acc 96.8750 (97.3438) lr 0.260000 +epoch: [138/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.4011 (1.3184) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 821.272, TIME@all 0.312 +epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:55:23 loss 1.1720 (1.2680) acc 100.0000 (98.2812) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:55:14 loss 1.2676 (1.2892) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 821.277, TIME@all 0.312 +epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.4243 (1.2878) acc 96.8750 (97.8125) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3302 (1.3097) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 821.329, TIME@all 0.312 +epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.4315 (1.2970) acc 96.8750 (96.8750) lr 0.260000 +epoch: [138/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3129 (1.3071) acc 96.8750 (96.3281) lr 0.260000 +FPS@all 821.367, TIME@all 0.312 +epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2118 (1.2753) acc 100.0000 (97.6562) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.2226 (1.3002) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 821.290, TIME@all 0.312 +epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2020 (1.2711) acc 100.0000 (98.5938) lr 0.260000 +epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.5243 (1.3016) acc 87.5000 (97.3438) lr 0.260000 +FPS@all 821.326, TIME@all 0.312 +epoch: [139/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:12 loss 1.5441 (1.3354) acc 93.7500 (96.0938) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5793 (1.3677) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 819.198, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:10 loss 1.4881 (1.3068) acc 96.8750 (97.5000) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5108 (1.3383) acc 90.6250 (96.6406) lr 0.260000 +FPS@all 819.298, TIME@all 0.312 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.3762 (1.3419) acc 96.8750 (96.5625) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.3505 (1.3769) acc 90.6250 (95.8594) lr 0.260000 +FPS@all 819.133, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:55:10 loss 1.4094 (1.3288) acc 96.8750 (96.2500) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:55:06 loss 1.4796 (1.3511) acc 93.7500 (96.0156) lr 0.260000 +FPS@all 819.148, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:55:11 loss 1.4845 (1.3660) acc 93.7500 (96.2500) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.4204 (1.3805) acc 93.7500 (95.6250) lr 0.260000 +FPS@all 819.154, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.5491 (1.3669) acc 93.7500 (95.7812) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.3355 (1.3779) acc 93.7500 (95.9375) lr 0.260000 +FPS@all 819.166, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:55:11 loss 1.4810 (1.3419) acc 96.8750 (96.4062) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:55:06 loss 1.3898 (1.3614) acc 100.0000 (95.9375) lr 0.260000 +FPS@all 819.183, TIME@all 0.313 +epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.4743 (1.3438) acc 96.8750 (95.9375) lr 0.260000 +epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5403 (1.3717) acc 90.6250 (95.6250) lr 0.260000 +FPS@all 819.174, TIME@all 0.313 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.4470 (1.3174) acc 96.8750 (96.7188) lr 0.260000 +epoch: [140/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:54:58 loss 1.2609 (1.3133) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 816.357, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3200 (1.2924) acc 96.8750 (97.6562) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:58 loss 1.2149 (1.2937) acc 100.0000 (97.4219) lr 0.260000 +FPS@all 816.422, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.3681 (1.3100) acc 93.7500 (97.0312) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.2295 (1.3138) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 816.266, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.3224 (1.2879) acc 96.8750 (97.8125) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3063 (1.3067) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 816.239, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3472 (1.2975) acc 96.8750 (97.3438) lr 0.260000 +epoch: [140/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3434 (1.3054) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 816.338, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.5682 (1.3206) acc 96.8750 (97.3438) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.2570 (1.3170) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 816.320, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3718 (1.2795) acc 96.8750 (98.1250) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3270 (1.2799) acc 93.7500 (97.9688) lr 0.260000 +FPS@all 816.329, TIME@all 0.314 +epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3783 (1.2846) acc 96.8750 (98.4375) lr 0.260000 +epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3377 (1.2966) acc 93.7500 (97.8125) lr 0.260000 +FPS@all 816.308, TIME@all 0.314 +epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.2899 (1.2886) acc 96.8750 (97.9688) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.3957 (1.3138) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 819.780, TIME@all 0.312 +epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3784 (1.2635) acc 96.8750 (98.2812) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4354 (1.2939) acc 93.7500 (98.0469) lr 0.260000 +FPS@all 819.631, TIME@all 0.312 +epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3668 (1.2953) acc 96.8750 (97.1875) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4462 (1.3207) acc 93.7500 (96.2500) lr 0.260000 +FPS@all 819.680, TIME@all 0.312 +epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3688 (1.2935) acc 96.8750 (97.6562) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:54:33 loss 1.3553 (1.3390) acc 96.8750 (96.0156) lr 0.260000 +FPS@all 819.643, TIME@all 0.312 +epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:54:50 loss 1.3202 (1.2968) acc 96.8750 (97.8125) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:54:33 loss 1.8302 (1.3142) acc 84.3750 (97.1094) lr 0.260000 +FPS@all 819.626, TIME@all 0.312 +epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.4627 (1.2750) acc 93.7500 (98.4375) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4423 (1.3153) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 819.654, TIME@all 0.312 +epoch: [141/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3009 (1.2794) acc 96.8750 (96.7188) lr 0.260000 +epoch: [141/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.5243 (1.3163) acc 90.6250 (96.3281) lr 0.260000 +FPS@all 819.638, TIME@all 0.312 +epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.4179 (1.2853) acc 96.8750 (97.0312) lr 0.260000 +epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.3025 (1.3102) acc 96.8750 (96.5625) lr 0.260000 +FPS@all 819.645, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.3344 (1.3175) acc 96.8750 (96.5625) lr 0.260000 +epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2469 (1.3253) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 821.526, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.3265 (1.3308) acc 93.7500 (96.5625) lr 0.260000 +epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2549 (1.3336) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 821.552, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.2971 (1.3132) acc 96.8750 (96.4062) lr 0.260000 +epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2757 (1.3272) acc 93.7500 (95.8594) lr 0.260000 +FPS@all 821.492, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:54:24 loss 1.4382 (1.3251) acc 93.7500 (97.1875) lr 0.260000 +epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:54:10 loss 1.2840 (1.3243) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 821.433, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:54:24 loss 1.4371 (1.3341) acc 96.8750 (97.0312) lr 0.260000 +epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:54:10 loss 1.3809 (1.3330) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 821.438, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.3842 (1.3421) acc 90.6250 (95.7812) lr 0.260000 +epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.3971 (1.3238) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 821.462, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.2558 (1.3138) acc 100.0000 (97.9688) lr 0.260000 +epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2348 (1.3293) acc 100.0000 (96.7969) lr 0.260000 +FPS@all 821.457, TIME@all 0.312 +epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.3104 (1.3142) acc 96.8750 (97.5000) lr 0.260000 +epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2873 (1.3235) acc 96.8750 (97.1875) lr 0.260000 +FPS@all 821.474, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3324 (1.2590) acc 96.8750 (98.1250) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2632 (1.2771) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 819.891, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:10 loss 1.3047 (1.2866) acc 93.7500 (97.5000) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2220 (1.3007) acc 100.0000 (97.6562) lr 0.260000 +FPS@all 819.970, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.2664 (1.2763) acc 96.8750 (97.8125) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:01 loss 1.3104 (1.3095) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 819.785, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:54:11 loss 1.2360 (1.2685) acc 100.0000 (97.9688) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:54:01 loss 1.3476 (1.3002) acc 93.7500 (97.2656) lr 0.260000 +FPS@all 819.804, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3154 (1.2767) acc 93.7500 (97.9688) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2119 (1.2934) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 819.829, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:54:11 loss 1.3129 (1.2645) acc 100.0000 (98.2812) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2200 (1.2885) acc 100.0000 (96.9531) lr 0.260000 +FPS@all 819.837, TIME@all 0.312 +epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3177 (1.2878) acc 100.0000 (97.6562) lr 0.260000 +epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2783 (1.2949) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 819.851, TIME@all 0.312 +epoch: [143/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:54:11 loss 1.2512 (1.2580) acc 100.0000 (97.8125) lr 0.260000 +epoch: [143/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:54:01 loss 1.2167 (1.3013) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 819.846, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:53:51 loss 1.2800 (1.2813) acc 96.8750 (97.0312) lr 0.260000 +epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3164 (1.3073) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 819.560, TIME@all 0.312 +epoch: [144/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.1827 (1.2746) acc 100.0000 (97.6562) lr 0.260000 +epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3127 (1.3066) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 819.515, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:52 loss 1.3411 (1.2912) acc 96.8750 (97.0312) lr 0.260000 +epoch: [144/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:53:41 loss 1.3074 (1.3049) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 819.395, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:53:52 loss 1.2172 (1.2805) acc 96.8750 (96.4062) lr 0.260000 +epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:53:41 loss 1.3115 (1.3032) acc 96.8750 (96.8750) lr 0.260000 +FPS@all 819.413, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:53:52 loss 1.2686 (1.2754) acc 93.7500 (96.4062) lr 0.260000 +epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2814 (1.2904) acc 100.0000 (96.6406) lr 0.260000 +FPS@all 819.456, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.1580 (1.2807) acc 100.0000 (97.8125) lr 0.260000 +epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3171 (1.3010) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 819.444, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.3047 (1.2778) acc 100.0000 (97.8125) lr 0.260000 +epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2144 (1.2989) acc 100.0000 (97.1094) lr 0.260000 +FPS@all 819.506, TIME@all 0.312 +epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:52 loss 1.2027 (1.2650) acc 100.0000 (97.5000) lr 0.260000 +epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2080 (1.2904) acc 96.8750 (96.7969) lr 0.260000 +FPS@all 819.464, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3357 (1.2578) acc 96.8750 (97.9688) lr 0.260000 +epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.4115 (1.2876) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 821.037, TIME@all 0.312 +epoch: [145/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:53:31 loss 1.1971 (1.2327) acc 100.0000 (98.4375) lr 0.260000 +epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:24 loss 1.3621 (1.2683) acc 96.8750 (97.4219) lr 0.260000 +FPS@all 821.054, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:32 loss 1.2022 (1.2720) acc 100.0000 (98.1250) lr 0.260000 +epoch: [145/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:53:25 loss 1.4300 (1.2932) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 820.903, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.2889 (1.2500) acc 96.8750 (97.8125) lr 0.260000 +epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.6729 (1.2756) acc 87.5000 (97.4219) lr 0.260000 +FPS@all 820.983, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.001 (0.013) eta 0:53:32 loss 1.2795 (1.2651) acc 93.7500 (97.3438) lr 0.260000 +epoch: [145/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:53:24 loss 1.6089 (1.2895) acc 84.3750 (97.3438) lr 0.260000 +FPS@all 820.934, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3493 (1.2587) acc 90.6250 (97.1875) lr 0.260000 +epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.4169 (1.2770) acc 96.8750 (97.5781) lr 0.260000 +FPS@all 820.918, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.4527 (1.2534) acc 93.7500 (98.1250) lr 0.260000 +epoch: [145/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:53:24 loss 1.3442 (1.2835) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 821.003, TIME@all 0.312 +epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3707 (1.2722) acc 96.8750 (98.4375) lr 0.260000 +epoch: [145/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:53:24 loss 1.3266 (1.2810) acc 96.8750 (97.9688) lr 0.260000 +FPS@all 820.981, TIME@all 0.312 +epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:19 loss 1.2036 (1.2608) acc 100.0000 (97.5000) lr 0.260000 +epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.1901 (1.3021) acc 100.0000 (96.4062) lr 0.260000 +FPS@all 819.697, TIME@all 0.312 +epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:19 loss 1.4337 (1.2988) acc 93.7500 (97.5000) lr 0.260000 +epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2630 (1.3346) acc 96.8750 (96.4844) lr 0.260000 +FPS@all 819.687, TIME@all 0.312 +epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.3097 (1.2826) acc 100.0000 (97.8125) lr 0.260000 +epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2493 (1.3010) acc 93.7500 (97.4219) lr 0.260000 +FPS@all 819.552, TIME@all 0.312 +epoch: [146/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:53:20 loss 1.4727 (1.2797) acc 93.7500 (98.1250) lr 0.260000 +epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:53:15 loss 1.3330 (1.2966) acc 96.8750 (97.8125) lr 0.260000 +FPS@all 819.588, TIME@all 0.312 +epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.4380 (1.2984) acc 96.8750 (97.3438) lr 0.260000 +epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2718 (1.3137) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 819.648, TIME@all 0.312 +epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.5463 (1.2822) acc 87.5000 (98.1250) lr 0.260000 +epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2692 (1.3055) acc 96.8750 (97.6562) lr 0.260000 +FPS@all 819.623, TIME@all 0.312 +epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.2797 (1.2993) acc 93.7500 (96.8750) lr 0.260000 +epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.3690 (1.2995) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 819.601, TIME@all 0.312 +epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.3978 (1.3016) acc 93.7500 (96.8750) lr 0.260000 +epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.3959 (1.3105) acc 93.7500 (96.7188) lr 0.260000 +FPS@all 819.429, TIME@all 0.312 +epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.3507 (1.2709) acc 100.0000 (97.5000) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.2979 (1.3085) acc 96.8750 (97.1094) lr 0.260000 +FPS@all 819.343, TIME@all 0.312 +epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.3864 (1.2765) acc 96.8750 (98.2812) lr 0.260000 +epoch: [147/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.2503 (1.2850) acc 93.7500 (97.7344) lr 0.260000 +FPS@all 819.404, TIME@all 0.312 +epoch: [147/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:53:01 loss 1.3498 (1.2768) acc 93.7500 (97.9688) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:52:55 loss 1.2707 (1.2927) acc 96.8750 (97.2656) lr 0.260000 +FPS@all 819.234, TIME@all 0.312 +epoch: [147/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:53:01 loss 1.2052 (1.2894) acc 100.0000 (96.8750) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:52:55 loss 1.3789 (1.3051) acc 96.8750 (97.0312) lr 0.260000 +FPS@all 819.270, TIME@all 0.312 +epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.3381 (1.2547) acc 96.8750 (98.4375) lr 0.260000 +epoch: [147/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.2865 (1.2942) acc 93.7500 (97.0312) lr 0.260000 +FPS@all 819.342, TIME@all 0.312 +epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.2945 (1.2495) acc 96.8750 (98.2812) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.1901 (1.2903) acc 100.0000 (97.1875) lr 0.260000 +FPS@all 819.339, TIME@all 0.312 +epoch: [147/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.3316 (1.2722) acc 96.8750 (97.1875) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.3496 (1.2924) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 819.344, TIME@all 0.312 +epoch: [147/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.4370 (1.2872) acc 93.7500 (97.0312) lr 0.260000 +epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.2425 (1.3080) acc 100.0000 (96.7188) lr 0.260000 +FPS@all 819.294, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:52:40 loss 1.2006 (1.2685) acc 100.0000 (97.5000) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:35 loss 1.4368 (1.3026) acc 96.8750 (96.6406) lr 0.260000 +FPS@all 821.766, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.3127 (1.2767) acc 96.8750 (98.1250) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3369 (1.3154) acc 93.7500 (96.7969) lr 0.260000 +FPS@all 821.650, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.3071 (1.3023) acc 100.0000 (97.6562) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.4755 (1.3248) acc 90.6250 (96.9531) lr 0.260000 +FPS@all 821.582, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:42 loss 1.4044 (1.2865) acc 96.8750 (97.8125) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.5110 (1.3181) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 821.562, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4168 (1.2882) acc 90.6250 (97.6562) lr 0.260000 +epoch: [148/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3506 (1.3442) acc 93.7500 (96.6406) lr 0.260000 +FPS@all 821.661, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4677 (1.2808) acc 93.7500 (98.1250) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3924 (1.3149) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 821.581, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:52:41 loss 1.4685 (1.2995) acc 93.7500 (97.3438) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.5286 (1.3141) acc 90.6250 (97.0312) lr 0.260000 +FPS@all 821.606, TIME@all 0.312 +epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4934 (1.3033) acc 90.6250 (96.5625) lr 0.260000 +epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3794 (1.3334) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 821.605, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.3518 (1.3037) acc 96.8750 (97.0312) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.4412 (1.3119) acc 90.6250 (96.5625) lr 0.260000 +FPS@all 820.226, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:52:31 loss 1.2645 (1.2729) acc 100.0000 (97.5000) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:52:23 loss 1.4622 (1.2994) acc 93.7500 (97.1094) lr 0.260000 +FPS@all 820.133, TIME@all 0.312 +epoch: [149/350][20/50] time 0.309 (0.313) data 0.000 (0.014) eta 0:52:30 loss 1.2608 (1.2959) acc 96.8750 (97.5000) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3316 (1.2930) acc 93.7500 (97.3438) lr 0.260000 +FPS@all 820.234, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.2324 (1.3195) acc 96.8750 (96.8750) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3053 (1.3279) acc 100.0000 (96.4844) lr 0.260000 +FPS@all 820.151, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:31 loss 1.2632 (1.2636) acc 100.0000 (99.2188) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.2418 (1.2883) acc 100.0000 (98.2812) lr 0.260000 +FPS@all 820.176, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:31 loss 1.2376 (1.2727) acc 96.8750 (97.6562) lr 0.260000 +epoch: [149/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:52:23 loss 1.4074 (1.2908) acc 96.8750 (96.9531) lr 0.260000 +FPS@all 820.150, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.3735 (1.2807) acc 93.7500 (96.5625) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3156 (1.3001) acc 96.8750 (96.4062) lr 0.260000 +FPS@all 820.138, TIME@all 0.312 +epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.1852 (1.2862) acc 100.0000 (97.3438) lr 0.260000 +epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.2744 (1.2981) acc 100.0000 (97.5000) lr 0.260000 +FPS@all 820.182, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:13 loss 1.2652 (1.2369) acc 100.0000 (98.5938) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:10 loss 1.2839 (1.2688) acc 100.0000 (97.8906) lr 0.260000 +FPS@all 819.938, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3262 (1.2410) acc 96.8750 (98.5938) lr 0.260000 +epoch: [150/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3729 (1.3000) acc 93.7500 (96.9531) lr 0.260000 +FPS@all 819.821, TIME@all 0.312 +epoch: [150/350][20/50] time 0.306 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.5075 (1.2949) acc 90.6250 (97.1875) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3337 (1.3147) acc 93.7500 (96.3281) lr 0.260000 +FPS@all 819.727, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3045 (1.2773) acc 96.8750 (98.1250) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.4422 (1.3003) acc 93.7500 (97.1875) lr 0.260000 +FPS@all 819.800, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3580 (1.2617) acc 90.6250 (98.2812) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3714 (1.3012) acc 96.8750 (97.5000) lr 0.260000 +FPS@all 819.716, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3207 (1.2793) acc 100.0000 (97.5000) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.5099 (1.3022) acc 90.6250 (97.2656) lr 0.260000 +FPS@all 819.752, TIME@all 0.312 +epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.2934 (1.2575) acc 100.0000 (97.9688) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3508 (1.2848) acc 96.8750 (97.3438) lr 0.260000 +FPS@all 819.566, TIME@all 0.312 +epoch: [150/350][20/50] time 0.306 (0.313) data 0.000 (0.013) eta 0:52:14 loss 1.2708 (1.2654) acc 100.0000 (98.2812) lr 0.260000 +epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:52:11 loss 1.3161 (1.3001) acc 93.7500 (97.5781) lr 0.260000 +FPS@all 819.762, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1757 (1.2032) acc 100.0000 (98.7500) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1901 (1.2070) acc 100.0000 (98.6719) lr 0.026000 +FPS@all 819.772, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:51:55 loss 1.1401 (1.2007) acc 100.0000 (99.3750) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:50 loss 1.2448 (1.2158) acc 96.8750 (98.6719) lr 0.026000 +FPS@all 819.672, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:51:54 loss 1.2299 (1.2158) acc 96.8750 (98.5938) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.2319 (1.2227) acc 100.0000 (98.4375) lr 0.026000 +FPS@all 819.602, TIME@all 0.312 +epoch: [151/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:51:54 loss 1.1744 (1.1974) acc 100.0000 (99.6875) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.1756 (1.2157) acc 96.8750 (98.5938) lr 0.026000 +FPS@all 819.583, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:51:55 loss 1.1654 (1.2024) acc 96.8750 (98.9062) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.1466 (1.2036) acc 100.0000 (98.5938) lr 0.026000 +FPS@all 819.635, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:55 loss 1.1773 (1.2070) acc 100.0000 (99.3750) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:51 loss 1.1540 (1.2164) acc 100.0000 (98.7500) lr 0.026000 +FPS@all 819.616, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1788 (1.2123) acc 100.0000 (99.0625) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1278 (1.2023) acc 100.0000 (99.0625) lr 0.026000 +FPS@all 819.683, TIME@all 0.312 +epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1732 (1.2140) acc 100.0000 (98.7500) lr 0.026000 +epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1512 (1.2099) acc 100.0000 (98.5938) lr 0.026000 +FPS@all 819.672, TIME@all 0.312 +epoch: [152/350][20/50] time 0.319 (0.315) data 0.001 (0.014) eta 0:52:06 loss 1.1614 (1.1623) acc 100.0000 (99.5312) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1513 (1.1616) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 815.327, TIME@all 0.314 +epoch: [152/350][20/50] time 0.319 (0.315) data 0.000 (0.012) eta 0:52:06 loss 1.1728 (1.1479) acc 100.0000 (99.8438) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:52:04 loss 1.1456 (1.1587) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 815.184, TIME@all 0.314 +epoch: [152/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:52:06 loss 1.2145 (1.1579) acc 96.8750 (99.3750) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1283 (1.1551) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 815.185, TIME@all 0.314 +epoch: [152/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:52:06 loss 1.2327 (1.1468) acc 96.8750 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1239 (1.1442) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 815.249, TIME@all 0.314 +epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1400 (1.1514) acc 100.0000 (99.5312) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1615 (1.1568) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 815.239, TIME@all 0.314 +epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1561 (1.1541) acc 100.0000 (99.6875) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1175 (1.1533) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 815.190, TIME@all 0.314 +epoch: [152/350][20/50] time 0.319 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1795 (1.1524) acc 100.0000 (99.3750) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1390 (1.1513) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 815.208, TIME@all 0.314 +epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:05 loss 1.4165 (1.1698) acc 93.7500 (99.5312) lr 0.026000 +epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:03 loss 1.1470 (1.1663) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 815.362, TIME@all 0.314 +epoch: [153/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 0:51:25 loss 1.1228 (1.1245) acc 100.0000 (100.0000) lr 0.026000 +epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:19 loss 1.1777 (1.1345) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 821.441, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1448 (1.1319) acc 96.8750 (99.3750) lr 0.026000 +epoch: [153/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:51:20 loss 1.1731 (1.1454) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 821.278, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1854 (1.1363) acc 96.8750 (99.6875) lr 0.026000 +epoch: [153/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1384 (1.1390) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 821.305, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1107 (1.1264) acc 100.0000 (100.0000) lr 0.026000 +epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1159 (1.1394) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 821.315, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1369 (1.1244) acc 100.0000 (100.0000) lr 0.026000 +epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1371 (1.1439) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 821.319, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1170 (1.1328) acc 100.0000 (99.6875) lr 0.026000 +epoch: [153/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:51:20 loss 1.1460 (1.1408) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 821.327, TIME@all 0.312 +epoch: [153/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:51:26 loss 1.2284 (1.1283) acc 96.8750 (99.8438) lr 0.026000 +epoch: [153/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.2121 (1.1514) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 821.320, TIME@all 0.312 +epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1057 (1.1293) acc 100.0000 (99.5312) lr 0.026000 +epoch: [153/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 0:51:20 loss 1.1188 (1.1376) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 821.304, TIME@all 0.312 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:13 loss 1.1452 (1.1204) acc 100.0000 (100.0000) lr 0.026000 +epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:22 loss 1.2816 (1.1319) acc 93.7500 (99.5312) lr 0.026000 +FPS@all 816.662, TIME@all 0.313 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1021 (1.1252) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1223 (1.1285) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.452, TIME@all 0.314 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:13 loss 1.1103 (1.1200) acc 100.0000 (100.0000) lr 0.026000 +epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1593 (1.1250) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 816.529, TIME@all 0.314 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1408 (1.1135) acc 100.0000 (100.0000) lr 0.026000 +epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1953 (1.1315) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.432, TIME@all 0.314 +epoch: [154/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1309 (1.1179) acc 100.0000 (100.0000) lr 0.026000 +epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1856 (1.1357) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 816.517, TIME@all 0.314 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1020 (1.1141) acc 100.0000 (99.8438) lr 0.026000 +epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1251 (1.1313) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 816.512, TIME@all 0.314 +epoch: [154/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1552 (1.1240) acc 100.0000 (99.6875) lr 0.026000 +epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1250 (1.1293) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 816.470, TIME@all 0.314 +epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1164 (1.1211) acc 100.0000 (99.8438) lr 0.026000 +epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1266 (1.1269) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 816.553, TIME@all 0.314 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.2142 (1.1308) acc 96.8750 (99.3750) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1575 (1.1425) acc 100.0000 (99.1406) lr 0.026000 +FPS@all 819.568, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:50:55 loss 1.2218 (1.1237) acc 96.8750 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1317 (1.1385) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 819.645, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:56 loss 1.2237 (1.1162) acc 96.8750 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1232 (1.1280) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 819.478, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.1579 (1.1180) acc 100.0000 (100.0000) lr 0.026000 +epoch: [155/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.0979 (1.1407) acc 100.0000 (99.2969) lr 0.026000 +FPS@all 819.535, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.2122 (1.1182) acc 96.8750 (99.8438) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:50:54 loss 1.1190 (1.1343) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.554, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:56 loss 1.1750 (1.1279) acc 100.0000 (99.6875) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1074 (1.1410) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 819.525, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:50:56 loss 1.1491 (1.1253) acc 100.0000 (99.8438) lr 0.026000 +epoch: [155/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:50:54 loss 1.1308 (1.1407) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.471, TIME@all 0.312 +epoch: [155/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:50:56 loss 1.1679 (1.1121) acc 100.0000 (99.8438) lr 0.026000 +epoch: [155/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1229 (1.1315) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 819.494, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:37 loss 1.1063 (1.1156) acc 100.0000 (99.3750) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:50:35 loss 1.2050 (1.1293) acc 96.8750 (99.2188) lr 0.026000 +FPS@all 820.623, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.1373 (1.1091) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1184 (1.1207) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.564, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0857 (1.1049) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:36 loss 1.1245 (1.1233) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.429, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:37 loss 1.1230 (1.1117) acc 100.0000 (99.6875) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1255 (1.1317) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 820.547, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:50:38 loss 1.1414 (1.1042) acc 100.0000 (100.0000) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:50:36 loss 1.1018 (1.1167) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.476, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0908 (1.1091) acc 100.0000 (99.6875) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:36 loss 1.1560 (1.1243) acc 96.8750 (99.2969) lr 0.026000 +FPS@all 820.534, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0933 (1.1107) acc 100.0000 (99.6875) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1167 (1.1229) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.559, TIME@all 0.312 +epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0908 (1.1108) acc 100.0000 (99.8438) lr 0.026000 +epoch: [156/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:50:36 loss 1.1442 (1.1207) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.513, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1583 (1.1052) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.0972 (1.1151) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.911, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1054 (1.1075) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1198 (1.1135) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.802, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1080 (1.1073) acc 100.0000 (99.8438) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1094 (1.1169) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 820.813, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.0880 (1.0970) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1190 (1.1119) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.809, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:50:21 loss 1.1146 (1.1059) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:50:18 loss 1.1003 (1.1115) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.739, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1199 (1.1049) acc 100.0000 (99.6875) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.0996 (1.1072) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.808, TIME@all 0.312 +epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1047 (1.1038) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1022 (1.1164) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.775, TIME@all 0.312 +epoch: [157/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1010 (1.0985) acc 100.0000 (100.0000) lr 0.026000 +epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1010 (1.1164) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.814, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1387 (1.1080) acc 100.0000 (99.8438) lr 0.026000 +epoch: [158/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1457 (1.1161) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 820.871, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2187 (1.1095) acc 96.8750 (99.6875) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.1735 (1.1186) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.772, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1512 (1.1103) acc 100.0000 (99.6875) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1036 (1.1234) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 820.846, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2896 (1.1123) acc 93.7500 (99.6875) lr 0.026000 +epoch: [158/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.0897 (1.1172) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.715, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.1998 (1.1053) acc 100.0000 (100.0000) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.1093 (1.1105) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.765, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1775 (1.1082) acc 100.0000 (100.0000) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:01 loss 1.1041 (1.1160) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.801, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2224 (1.1104) acc 96.8750 (99.5312) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:01 loss 1.1017 (1.1183) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 820.817, TIME@all 0.312 +epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1557 (1.1040) acc 100.0000 (99.8438) lr 0.026000 +epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1157 (1.1190) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.814, TIME@all 0.312 +epoch: [159/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1042 (1.0983) acc 100.0000 (99.8438) lr 0.026000 +epoch: [159/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:49:47 loss 1.1180 (1.1091) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 820.505, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.1434 (1.1019) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:49:47 loss 1.1191 (1.1073) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.506, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1227 (1.0963) acc 100.0000 (99.8438) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1407 (1.1095) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.396, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1277 (1.1006) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1316 (1.1082) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 820.405, TIME@all 0.312 +epoch: [159/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:53 loss 1.1236 (1.1023) acc 100.0000 (99.8438) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1739 (1.1087) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.439, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.0924 (1.1039) acc 100.0000 (99.8438) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:47 loss 1.1499 (1.1067) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.445, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1167 (1.1010) acc 100.0000 (99.6875) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1396 (1.1055) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.424, TIME@all 0.312 +epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.0859 (1.1042) acc 100.0000 (100.0000) lr 0.026000 +epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:47 loss 1.1506 (1.1101) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.407, TIME@all 0.312 +epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0852 (1.0966) acc 100.0000 (99.6875) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0841 (1.1123) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 813.442, TIME@all 0.315 +epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0847 (1.1012) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0880 (1.1167) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 813.475, TIME@all 0.315 +epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.011) eta 0:50:29 loss 1.1003 (1.0911) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.311 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0948 (1.1092) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 813.532, TIME@all 0.315 +epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1033 (1.0975) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.0855 (1.1089) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 813.325, TIME@all 0.315 +epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0974 (1.0908) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.1063 (1.1045) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 813.392, TIME@all 0.315 +epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1040 (1.0936) acc 100.0000 (99.8438) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.001 (0.006) eta 0:50:04 loss 1.0997 (1.1131) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 813.405, TIME@all 0.315 +epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1075 (1.0975) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.0893 (1.1065) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 813.366, TIME@all 0.315 +epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0956 (1.0897) acc 100.0000 (100.0000) lr 0.026000 +epoch: [160/350][40/50] time 0.312 (0.316) data 0.001 (0.006) eta 0:50:04 loss 1.0797 (1.1041) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 813.373, TIME@all 0.315 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.013) eta 0:49:57 loss 1.0921 (1.1046) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.007) eta 0:50:22 loss 1.1008 (1.1196) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 805.963, TIME@all 0.318 +epoch: [161/350][20/50] time 0.334 (0.316) data 0.000 (0.013) eta 0:49:57 loss 1.0952 (1.0993) acc 100.0000 (99.5312) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.007) eta 0:50:21 loss 1.0895 (1.1082) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 806.035, TIME@all 0.318 +epoch: [161/350][20/50] time 0.334 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.1328 (1.0978) acc 100.0000 (100.0000) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.320) data 0.000 (0.006) eta 0:50:22 loss 1.0729 (1.1080) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 805.886, TIME@all 0.318 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:56 loss 1.0945 (1.0965) acc 100.0000 (100.0000) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.0996 (1.1074) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 806.148, TIME@all 0.318 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.001 (0.012) eta 0:49:57 loss 1.0965 (1.1041) acc 100.0000 (99.2188) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.0897 (1.1108) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 806.199, TIME@all 0.318 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.001 (0.012) eta 0:49:57 loss 1.1205 (1.1057) acc 100.0000 (99.3750) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.001 (0.006) eta 0:50:21 loss 1.1138 (1.1127) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 806.020, TIME@all 0.318 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.0876 (1.0977) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:22 loss 1.0879 (1.1084) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 805.946, TIME@all 0.318 +epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.0914 (1.0956) acc 100.0000 (99.8438) lr 0.026000 +epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.1010 (1.1103) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 806.027, TIME@all 0.318 +epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.0997 (1.0897) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1382 (1.1014) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.297, TIME@all 0.312 +epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:10 loss 1.2048 (1.1035) acc 100.0000 (99.6875) lr 0.026000 +epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1027 (1.1107) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.352, TIME@all 0.312 +epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.1960 (1.0974) acc 100.0000 (100.0000) lr 0.026000 +epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:07 loss 1.1064 (1.0968) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.154, TIME@all 0.313 +epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.0960 (1.0921) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1279 (1.1049) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.271, TIME@all 0.312 +epoch: [162/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:49:11 loss 1.1384 (1.1116) acc 96.8750 (99.3750) lr 0.026000 +epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1473 (1.1108) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.281, TIME@all 0.312 +epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:11 loss 1.1183 (1.0939) acc 100.0000 (100.0000) lr 0.026000 +epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1246 (1.1082) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.194, TIME@all 0.313 +epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:49:11 loss 1.1065 (1.0953) acc 100.0000 (99.8438) lr 0.026000 +epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.0991 (1.1007) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.198, TIME@all 0.313 +epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.1042 (1.0960) acc 100.0000 (100.0000) lr 0.026000 +epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1819 (1.1070) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.185, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:48:56 loss 1.1592 (1.1137) acc 100.0000 (99.6875) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:48:50 loss 1.1122 (1.1193) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.046, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:48:56 loss 1.0961 (1.0920) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:48:50 loss 1.1262 (1.1021) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.086, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1075 (1.0905) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1375 (1.1038) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.974, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1247 (1.1102) acc 100.0000 (99.5312) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0967 (1.1128) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 818.971, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.0978 (1.0943) acc 100.0000 (100.0000) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1109 (1.1132) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.015, TIME@all 0.313 +epoch: [163/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1438 (1.1087) acc 100.0000 (99.3750) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1043 (1.1147) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 818.984, TIME@all 0.313 +epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1187 (1.0998) acc 100.0000 (99.6875) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0872 (1.1045) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.026, TIME@all 0.313 +epoch: [163/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.0996 (1.0999) acc 100.0000 (99.6875) lr 0.026000 +epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0865 (1.1033) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.963, TIME@all 0.313 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:48:33 loss 1.1142 (1.0848) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.0996 (1.0964) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.958, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1318 (1.0932) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.2026 (1.1110) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 820.773, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:48:33 loss 1.2138 (1.1032) acc 96.8750 (99.6875) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.0983 (1.1075) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.871, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:48:34 loss 1.1922 (1.0897) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:48:27 loss 1.0885 (1.1084) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 820.778, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1330 (1.0920) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1383 (1.1084) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.836, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:33 loss 1.1036 (1.0885) acc 100.0000 (100.0000) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1250 (1.1035) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.876, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:33 loss 1.1117 (1.0925) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1515 (1.1060) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.887, TIME@all 0.312 +epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1593 (1.1019) acc 100.0000 (99.8438) lr 0.026000 +epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1226 (1.1087) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.797, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.2141 (1.0954) acc 100.0000 (99.8438) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:48:13 loss 1.1854 (1.1004) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 820.488, TIME@all 0.312 +epoch: [165/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:48:17 loss 1.2252 (1.1000) acc 96.8750 (99.8438) lr 0.026000 +epoch: [165/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:48:13 loss 1.0970 (1.1021) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.530, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.2637 (1.1052) acc 100.0000 (99.6875) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:14 loss 1.1583 (1.1130) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 820.361, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.2306 (1.0983) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:13 loss 1.1546 (1.1030) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.411, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.1858 (1.0975) acc 96.8750 (99.6875) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:48:13 loss 1.1369 (1.1086) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.367, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.1564 (1.1072) acc 96.8750 (99.6875) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:48:13 loss 1.1066 (1.1085) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.429, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.2006 (1.0953) acc 100.0000 (99.8438) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:48:13 loss 1.1920 (1.1070) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.441, TIME@all 0.312 +epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.1919 (1.0915) acc 100.0000 (100.0000) lr 0.026000 +epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:13 loss 1.1109 (1.0990) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 820.426, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:48:12 loss 1.1346 (1.0974) acc 100.0000 (99.8438) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1516 (1.1135) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.617, TIME@all 0.312 +epoch: [166/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:48:11 loss 1.1108 (1.0892) acc 100.0000 (100.0000) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:47:58 loss 1.0916 (1.1113) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.698, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:48:12 loss 1.1281 (1.0945) acc 96.8750 (99.6875) lr 0.026000 +epoch: [166/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.0862 (1.1044) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.560, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:48:12 loss 1.1439 (1.0873) acc 100.0000 (100.0000) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:59 loss 1.0924 (1.1007) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.506, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:48:12 loss 1.1028 (1.0991) acc 100.0000 (99.6875) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1913 (1.1073) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 820.598, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:48:12 loss 1.1840 (1.1064) acc 96.8750 (99.5312) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1677 (1.1119) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 820.544, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:48:12 loss 1.1997 (1.0962) acc 96.8750 (99.6875) lr 0.026000 +epoch: [166/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:47:58 loss 1.0705 (1.1105) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 820.542, TIME@all 0.312 +epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:48:12 loss 1.1511 (1.0983) acc 100.0000 (100.0000) lr 0.026000 +epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1043 (1.1112) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.557, TIME@all 0.312 +epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:47:49 loss 1.0973 (1.0916) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0940 (1.1062) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.604, TIME@all 0.312 +epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0856 (1.0875) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:47:44 loss 1.0758 (1.0963) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.681, TIME@all 0.312 +epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.1391 (1.0911) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1359 (1.1043) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.439, TIME@all 0.312 +epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.1176 (1.0918) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1370 (1.1017) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.496, TIME@all 0.312 +epoch: [167/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:47:49 loss 1.0893 (1.0890) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0972 (1.0984) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.587, TIME@all 0.312 +epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.0700 (1.0930) acc 100.0000 (99.8438) lr 0.026000 +epoch: [167/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1173 (1.1046) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.534, TIME@all 0.312 +epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0865 (1.0916) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0852 (1.0998) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.532, TIME@all 0.312 +epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0713 (1.0870) acc 100.0000 (100.0000) lr 0.026000 +epoch: [167/350][40/50] time 0.316 (0.313) data 0.001 (0.007) eta 0:47:45 loss 1.0869 (1.1023) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.583, TIME@all 0.312 +epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1027 (1.0957) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0798 (1.1076) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 818.812, TIME@all 0.313 +epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.014) eta 0:47:46 loss 1.1694 (1.0993) acc 100.0000 (100.0000) lr 0.026000 +epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.2112 (1.1152) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 818.869, TIME@all 0.313 +epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.0752 (1.0906) acc 100.0000 (100.0000) lr 0.026000 +epoch: [168/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:47:35 loss 1.0830 (1.1016) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.720, TIME@all 0.313 +epoch: [168/350][20/50] time 0.313 (0.314) data 0.001 (0.014) eta 0:47:47 loss 1.1385 (1.0855) acc 96.8750 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0720 (1.0980) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.751, TIME@all 0.313 +epoch: [168/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1324 (1.0977) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.1014 (1.0996) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.731, TIME@all 0.313 +epoch: [168/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:46 loss 1.1331 (1.0927) acc 100.0000 (99.6875) lr 0.026000 +epoch: [168/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0698 (1.0966) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.778, TIME@all 0.313 +epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1511 (1.0997) acc 100.0000 (99.5312) lr 0.026000 +epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.1283 (1.1067) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 818.810, TIME@all 0.313 +epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.0797 (1.0908) acc 100.0000 (99.8438) lr 0.026000 +epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0974 (1.1011) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.819, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0705 (1.0833) acc 100.0000 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.1256 (1.0956) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.396, TIME@all 0.313 +epoch: [169/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:47:27 loss 1.0937 (1.0842) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0926 (1.0932) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 818.457, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:47:27 loss 1.1603 (1.0888) acc 96.8750 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:19 loss 1.1415 (1.0995) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 818.521, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:28 loss 1.0879 (1.0807) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:47:20 loss 1.1288 (1.0971) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.291, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0992 (1.0893) acc 100.0000 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0722 (1.1010) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.395, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0687 (1.0865) acc 100.0000 (99.8438) lr 0.026000 +epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0996 (1.0990) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.358, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:28 loss 1.1063 (1.0836) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.1410 (1.1017) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 818.347, TIME@all 0.313 +epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:47:28 loss 1.0964 (1.0828) acc 100.0000 (100.0000) lr 0.026000 +epoch: [169/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:47:20 loss 1.0907 (1.0918) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 818.321, TIME@all 0.313 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:02 loss 1.0937 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:56 loss 1.0953 (1.0959) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.663, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:02 loss 1.1135 (1.0802) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:57 loss 1.1537 (1.0966) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.561, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:47:03 loss 1.1011 (1.0822) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:46:57 loss 1.0950 (1.1013) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.500, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:47:03 loss 1.1051 (1.0851) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:46:57 loss 1.0788 (1.0976) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.477, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1104 (1.0899) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:57 loss 1.1646 (1.1015) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.558, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1073 (1.0843) acc 100.0000 (100.0000) lr 0.026000 +epoch: [170/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 0:46:57 loss 1.1065 (1.1019) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.523, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1341 (1.0893) acc 96.8750 (99.6875) lr 0.026000 +epoch: [170/350][40/50] time 0.318 (0.313) data 0.001 (0.006) eta 0:46:57 loss 1.1000 (1.0996) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.529, TIME@all 0.312 +epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:03 loss 1.1169 (1.0876) acc 100.0000 (99.8438) lr 0.026000 +epoch: [170/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 0:46:57 loss 1.0914 (1.0996) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.596, TIME@all 0.312 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1229 (1.0882) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0995 (1.1024) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.332, TIME@all 0.313 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:00 loss 1.1308 (1.0905) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0990 (1.1070) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.394, TIME@all 0.313 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.0923 (1.0846) acc 100.0000 (100.0000) lr 0.026000 +epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0714 (1.1027) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.230, TIME@all 0.313 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.0838 (1.0913) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.1146 (1.1066) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.264, TIME@all 0.313 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1385 (1.0885) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0776 (1.1052) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 818.295, TIME@all 0.313 +epoch: [171/350][20/50] time 0.313 (0.314) data 0.001 (0.012) eta 0:47:01 loss 1.1733 (1.0877) acc 100.0000 (100.0000) lr 0.026000 +epoch: [171/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 0:46:47 loss 1.0868 (1.1044) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 818.315, TIME@all 0.313 +epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:01 loss 1.1740 (1.0959) acc 100.0000 (99.8438) lr 0.026000 +epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:46:47 loss 1.0869 (1.1154) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.283, TIME@all 0.313 +epoch: [171/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1515 (1.0951) acc 96.8750 (99.6875) lr 0.026000 +epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0845 (1.1131) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 818.279, TIME@all 0.313 +epoch: [172/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0968 (1.0907) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:46:24 loss 1.0998 (1.0977) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 820.568, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1791 (1.0930) acc 96.8750 (99.6875) lr 0.026000 +epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1489 (1.0997) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.437, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0867 (1.0841) acc 100.0000 (99.8438) lr 0.026000 +epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:46:24 loss 1.0962 (1.0954) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.486, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1160 (1.0914) acc 100.0000 (99.8438) lr 0.026000 +epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1288 (1.1008) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.463, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:29 loss 1.1294 (1.0913) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1186 (1.1027) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.333, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0961 (1.0935) acc 100.0000 (99.6875) lr 0.026000 +epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1098 (1.0985) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.414, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:46:28 loss 1.0908 (1.0879) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1095 (1.0998) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.449, TIME@all 0.312 +epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1324 (1.0869) acc 100.0000 (100.0000) lr 0.026000 +epoch: [172/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:46:24 loss 1.1002 (1.1024) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.504, TIME@all 0.312 +epoch: [173/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:46:13 loss 1.0876 (1.0781) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1310 (1.0955) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.870, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1129 (1.0868) acc 96.8750 (99.5312) lr 0.026000 +epoch: [173/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1593 (1.0990) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 818.823, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1344 (1.0817) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:46:13 loss 1.2195 (1.1031) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 818.736, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1469 (1.0876) acc 100.0000 (99.8438) lr 0.026000 +epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1992 (1.1056) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 818.710, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:46:13 loss 1.1506 (1.0815) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:46:13 loss 1.1593 (1.0989) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 818.790, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1098 (1.0833) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1231 (1.1005) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.724, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.2137 (1.0838) acc 96.8750 (99.8438) lr 0.026000 +epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1481 (1.0952) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 818.770, TIME@all 0.313 +epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1142 (1.0846) acc 100.0000 (100.0000) lr 0.026000 +epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.0898 (1.0961) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.780, TIME@all 0.313 +epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:00 loss 1.0744 (1.0829) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:52 loss 1.2057 (1.1027) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 820.797, TIME@all 0.312 +epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0790 (1.0850) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:45:53 loss 1.1632 (1.0992) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.602, TIME@all 0.312 +epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:46:00 loss 1.0787 (1.0848) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:52 loss 1.1333 (1.0954) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 820.725, TIME@all 0.312 +epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0685 (1.0851) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.1209 (1.0969) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.688, TIME@all 0.312 +epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:46:00 loss 1.0823 (1.0853) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.1730 (1.1015) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 820.646, TIME@all 0.312 +epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0851 (1.0780) acc 100.0000 (99.8438) lr 0.026000 +epoch: [174/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:45:52 loss 1.0951 (1.0916) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.671, TIME@all 0.312 +epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.1121 (1.0833) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:53 loss 1.0994 (1.0961) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.634, TIME@all 0.312 +epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:46:01 loss 1.1133 (1.0836) acc 100.0000 (100.0000) lr 0.026000 +epoch: [174/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.0871 (1.0953) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.685, TIME@all 0.312 +epoch: [175/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0872 (1.0789) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0642 (1.0897) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.142, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0821 (1.0812) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0927 (1.0920) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 816.163, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:45:59 loss 1.0971 (1.0806) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:54 loss 1.0799 (1.0912) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.051, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:45:59 loss 1.0854 (1.0818) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.315) data 0.000 (0.006) eta 0:45:55 loss 1.0846 (1.0895) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 816.003, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0808 (1.0826) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0984 (1.0857) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 816.081, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0631 (1.0815) acc 100.0000 (99.6875) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0813 (1.0941) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 816.052, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0745 (1.0800) acc 100.0000 (99.8438) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:54 loss 1.0762 (1.0890) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 816.075, TIME@all 0.314 +epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0924 (1.0766) acc 100.0000 (100.0000) lr 0.026000 +epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0700 (1.0923) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.057, TIME@all 0.314 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.014) eta 0:45:38 loss 1.0942 (1.0837) acc 100.0000 (99.5312) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1388 (1.0942) acc 100.0000 (99.6875) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.733, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1086 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0991 (1.0995) acc 100.0000 (99.6875) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.776, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0848 (1.0798) acc 100.0000 (99.8438) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.2606 (1.0932) acc 96.8750 (99.7656) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.616, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1308 (1.0925) acc 96.8750 (99.6875) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:32 loss 1.0989 (1.0997) acc 100.0000 (99.6094) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.634, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1003 (1.0937) acc 100.0000 (99.5312) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1113 (1.0985) acc 100.0000 (99.7656) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.673, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1042 (1.0808) acc 100.0000 (100.0000) lr 0.026000 +epoch: [176/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0915 (1.0976) acc 100.0000 (99.8438) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.753, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0926 (1.0853) acc 100.0000 (99.8438) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0981 (1.0979) acc 100.0000 (99.8438) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.657, TIME@all 0.313 +epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0752 (1.0847) acc 100.0000 (100.0000) lr 0.026000 +epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1506 (1.0980) acc 100.0000 (99.9219) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 818.718, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1169 (1.0868) acc 100.0000 (99.6875) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:12 loss 1.0959 (1.1008) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.280, TIME@all 0.312 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.0809 (1.0811) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:13 loss 1.1460 (1.0998) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.122, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:17 loss 1.0847 (1.0983) acc 100.0000 (99.3750) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:12 loss 1.1162 (1.1003) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.298, TIME@all 0.312 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:45:18 loss 1.0858 (1.0868) acc 100.0000 (99.8438) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:45:13 loss 1.0993 (1.0981) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.130, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1589 (1.0851) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:45:12 loss 1.1792 (1.0994) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 819.198, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.2168 (1.0856) acc 96.8750 (99.6875) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:45:13 loss 1.1056 (1.1010) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.124, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1494 (1.0851) acc 100.0000 (99.8438) lr 0.026000 +epoch: [177/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:45:13 loss 1.1320 (1.0928) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.172, TIME@all 0.313 +epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.0915 (1.0764) acc 100.0000 (100.0000) lr 0.026000 +epoch: [177/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:45:13 loss 1.0856 (1.0897) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.169, TIME@all 0.313 +epoch: [178/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:45:00 loss 1.1846 (1.0917) acc 100.0000 (99.6875) lr 0.026000 +epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0805 (1.1035) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.257, TIME@all 0.312 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:44:59 loss 1.1686 (1.0946) acc 100.0000 (99.5312) lr 0.026000 +epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1024 (1.1070) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.312, TIME@all 0.312 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0808 (1.0880) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0874 (1.0989) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.114, TIME@all 0.313 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0755 (1.0885) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1078 (1.1000) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.157, TIME@all 0.313 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0919 (1.0843) acc 100.0000 (100.0000) lr 0.026000 +epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0657 (1.0916) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.226, TIME@all 0.312 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1188 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0871 (1.0958) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.232, TIME@all 0.312 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1162 (1.0908) acc 100.0000 (99.6875) lr 0.026000 +epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0921 (1.0961) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.216, TIME@all 0.312 +epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1376 (1.0833) acc 100.0000 (99.8438) lr 0.026000 +epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1025 (1.1024) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.220, TIME@all 0.312 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.014) eta 0:44:59 loss 1.1091 (1.0826) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.0947 (1.0883) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 815.478, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.0868 (1.0776) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1072 (1.0961) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 815.406, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:45:00 loss 1.0849 (1.0859) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1382 (1.0966) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 815.301, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1089 (1.0862) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.316 (0.315) data 0.001 (0.007) eta 0:44:57 loss 1.1127 (1.0929) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 815.352, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1143 (1.0760) acc 96.8750 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.2115 (1.0898) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 815.336, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.0968 (1.0783) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.2125 (1.0970) acc 93.7500 (99.8438) lr 0.026000 +FPS@all 815.393, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1363 (1.0831) acc 100.0000 (99.8438) lr 0.026000 +epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.0930 (1.0944) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 815.390, TIME@all 0.314 +epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.014) eta 0:44:59 loss 1.0902 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [179/350][40/50] time 0.316 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1532 (1.0908) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 815.375, TIME@all 0.314 +epoch: [180/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:44:25 loss 1.1041 (1.0857) acc 100.0000 (100.0000) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:19 loss 1.0855 (1.0986) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.558, TIME@all 0.312 +epoch: [180/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:44:25 loss 1.2310 (1.0939) acc 96.8750 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0602 (1.1018) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.433, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:26 loss 1.2113 (1.0956) acc 96.8750 (99.5312) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0845 (1.0998) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.291, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:44:25 loss 1.1287 (1.0869) acc 100.0000 (99.8438) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.1811 (1.0990) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 820.345, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.0828 (1.0895) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.1393 (1.1009) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.395, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.2384 (1.0958) acc 96.8750 (99.8438) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0908 (1.1009) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.388, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:44:25 loss 1.0874 (1.0851) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0952 (1.0984) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.387, TIME@all 0.312 +epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.1219 (1.0877) acc 100.0000 (99.6875) lr 0.026000 +epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0853 (1.0913) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.353, TIME@all 0.312 +epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.0907 (1.0975) acc 100.0000 (99.6875) lr 0.026000 +epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1287 (1.0981) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.391, TIME@all 0.313 +epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:44:23 loss 1.1234 (1.0879) acc 100.0000 (100.0000) lr 0.026000 +epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1072 (1.0991) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.404, TIME@all 0.313 +epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:44:24 loss 1.0822 (1.0906) acc 100.0000 (99.6875) lr 0.026000 +epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:44:17 loss 1.0925 (1.0969) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.235, TIME@all 0.313 +epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:44:24 loss 1.0653 (1.0933) acc 100.0000 (99.6875) lr 0.026000 +epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:17 loss 1.0901 (1.0946) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 817.284, TIME@all 0.313 +epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.0945 (1.0923) acc 100.0000 (99.8438) lr 0.026000 +epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1381 (1.1011) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 817.360, TIME@all 0.313 +epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:24 loss 1.0801 (1.0940) acc 100.0000 (99.5312) lr 0.026000 +epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:17 loss 1.1612 (1.0940) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 817.308, TIME@all 0.313 +epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.1109 (1.0932) acc 100.0000 (99.8438) lr 0.026000 +epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1168 (1.0966) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 817.348, TIME@all 0.313 +epoch: [181/350][20/50] time 0.316 (0.314) data 0.001 (0.013) eta 0:44:24 loss 1.0730 (1.0886) acc 100.0000 (100.0000) lr 0.026000 +epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.0738 (1.0981) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 817.319, TIME@all 0.313 +epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:44:01 loss 1.1048 (1.0785) acc 100.0000 (99.6875) lr 0.026000 +epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.1112 (1.0877) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.036, TIME@all 0.313 +epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:44:00 loss 1.0998 (1.0780) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.1174 (1.0841) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.083, TIME@all 0.313 +epoch: [182/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.1045 (1.0744) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1242 (1.0860) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 818.974, TIME@all 0.313 +epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.0811 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1171 (1.0840) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.954, TIME@all 0.313 +epoch: [182/350][20/50] time 0.319 (0.313) data 0.001 (0.012) eta 0:44:01 loss 1.0858 (1.0793) acc 100.0000 (100.0000) lr 0.026000 +epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1780 (1.0893) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.042, TIME@all 0.313 +epoch: [182/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 0:44:01 loss 1.0974 (1.0836) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.0862 (1.0894) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.933, TIME@all 0.313 +epoch: [182/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.0696 (1.0736) acc 100.0000 (99.8438) lr 0.026000 +epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1595 (1.0895) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.037, TIME@all 0.313 +epoch: [182/350][20/50] time 0.319 (0.313) data 0.001 (0.013) eta 0:44:00 loss 1.2617 (1.0860) acc 96.8750 (99.6875) lr 0.026000 +epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.0929 (1.0882) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.049, TIME@all 0.313 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:34 loss 1.0841 (1.0883) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:43:31 loss 1.1061 (1.0936) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.674, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:34 loss 1.0882 (1.0909) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:43:31 loss 1.1100 (1.0982) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.722, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1480 (1.0976) acc 100.0000 (99.5312) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:32 loss 1.0966 (1.0996) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.528, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.0924 (1.0894) acc 100.0000 (99.8438) lr 0.026000 +epoch: [183/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.2214 (1.1010) acc 93.7500 (99.6875) lr 0.026000 +FPS@all 819.606, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1171 (1.0787) acc 100.0000 (100.0000) lr 0.026000 +epoch: [183/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.0845 (1.0866) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.578, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:35 loss 1.1461 (1.0857) acc 96.8750 (99.8438) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:43:31 loss 1.1166 (1.0934) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.586, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1418 (1.1041) acc 100.0000 (99.6875) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.1194 (1.1032) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.633, TIME@all 0.312 +epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1208 (1.0948) acc 100.0000 (99.6875) lr 0.026000 +epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.0839 (1.1015) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.650, TIME@all 0.312 +epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:28 loss 1.0964 (1.0773) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:43:20 loss 1.0848 (1.0877) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.055, TIME@all 0.312 +epoch: [184/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:43:29 loss 1.0755 (1.0793) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:43:21 loss 1.1577 (1.0928) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.898, TIME@all 0.312 +epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0704 (1.0772) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0929 (1.0895) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.948, TIME@all 0.312 +epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:28 loss 1.0958 (1.0834) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:20 loss 1.1336 (1.0943) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.975, TIME@all 0.312 +epoch: [184/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0695 (1.0778) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.1108 (1.0938) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.943, TIME@all 0.312 +epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0921 (1.0801) acc 100.0000 (100.0000) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:43:21 loss 1.1255 (1.0896) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.856, TIME@all 0.312 +epoch: [184/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0665 (1.0783) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0920 (1.0896) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.942, TIME@all 0.312 +epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0706 (1.0816) acc 100.0000 (99.8438) lr 0.026000 +epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0795 (1.0876) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.946, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.014) eta 0:43:09 loss 1.1366 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0602 (1.0937) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.430, TIME@all 0.312 +epoch: [185/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1594 (1.0780) acc 96.8750 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0764 (1.0901) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.276, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:43:09 loss 1.1164 (1.0807) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:43:02 loss 1.0860 (1.0936) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.278, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.0793 (1.0821) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0794 (1.0978) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 820.365, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1308 (1.0862) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0747 (1.1008) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.292, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.0801 (1.0837) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:43:02 loss 1.0686 (1.0948) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.279, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:10 loss 1.0711 (1.0796) acc 100.0000 (99.8438) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0605 (1.0915) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.277, TIME@all 0.312 +epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1527 (1.0813) acc 100.0000 (100.0000) lr 0.026000 +epoch: [185/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:43:02 loss 1.0653 (1.0933) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.283, TIME@all 0.312 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:56 loss 1.1150 (1.0760) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:42:50 loss 1.0780 (1.0868) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.958, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.1010 (1.0841) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0701 (1.0926) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 818.829, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0937 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0705 (1.0839) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.823, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0944 (1.0742) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0566 (1.0837) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.849, TIME@all 0.313 +epoch: [186/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0973 (1.0732) acc 100.0000 (99.8438) lr 0.026000 +epoch: [186/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0660 (1.0851) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.845, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:42:57 loss 1.0899 (1.0726) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:42:50 loss 1.0827 (1.0802) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.870, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.1177 (1.0755) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0818 (1.0832) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.916, TIME@all 0.313 +epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0888 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0815 (1.0873) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.894, TIME@all 0.313 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:38 loss 1.1360 (1.0895) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.309 (0.313) data 0.001 (0.007) eta 0:42:31 loss 1.2239 (1.0915) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.929, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.312) data 0.000 (0.014) eta 0:42:34 loss 1.0961 (1.0832) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:30 loss 1.1490 (1.0943) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.329, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0764 (1.0802) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1886 (1.0893) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 819.769, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.1471 (1.0824) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.1412 (1.0920) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.750, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0988 (1.0844) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1115 (1.0934) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.758, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:42:39 loss 1.1327 (1.0786) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1840 (1.0922) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.810, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0997 (1.0835) acc 100.0000 (99.8438) lr 0.026000 +epoch: [187/350][40/50] time 0.309 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.0744 (1.0900) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.827, TIME@all 0.312 +epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.014) eta 0:42:39 loss 1.1215 (1.0898) acc 100.0000 (100.0000) lr 0.026000 +epoch: [187/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.1662 (1.0909) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.799, TIME@all 0.312 +epoch: [188/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.1162 (1.0744) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.0871 (1.0904) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 821.125, TIME@all 0.312 +epoch: [188/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.0874 (1.0717) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:12 loss 1.0756 (1.0835) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.168, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.0838 (1.0745) acc 100.0000 (99.8438) lr 0.026000 +epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.0616 (1.0912) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 821.033, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:18 loss 1.0719 (1.0737) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0901 (1.0862) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 821.120, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:19 loss 1.0698 (1.0732) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0734 (1.0854) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.063, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:19 loss 1.1190 (1.0783) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0709 (1.0858) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.062, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:18 loss 1.0813 (1.0835) acc 100.0000 (99.8438) lr 0.026000 +epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.1038 (1.0977) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 821.113, TIME@all 0.312 +epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:18 loss 1.1076 (1.0761) acc 100.0000 (100.0000) lr 0.026000 +epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.1010 (1.0855) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.113, TIME@all 0.312 +epoch: [189/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1836 (1.0847) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:42:02 loss 1.0671 (1.0933) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.084, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:42:11 loss 1.1868 (1.0739) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0645 (1.0922) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.969, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.1124 (1.0800) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0681 (1.0967) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.964, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1013 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:42:03 loss 1.0829 (1.0925) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.031, TIME@all 0.313 +epoch: [189/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1601 (1.0769) acc 100.0000 (100.0000) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.1035 (1.0878) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.990, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.0813 (1.0781) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0680 (1.0907) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.967, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.001 (0.013) eta 0:42:11 loss 1.0770 (1.0785) acc 100.0000 (99.6875) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:42:03 loss 1.0708 (1.0885) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.971, TIME@all 0.313 +epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.0991 (1.0824) acc 100.0000 (99.8438) lr 0.026000 +epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0802 (1.0936) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.013, TIME@all 0.313 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0831 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.0850 (1.0822) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.411, TIME@all 0.312 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0711 (1.0693) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1006 (1.0814) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 820.464, TIME@all 0.312 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:41:54 loss 1.0891 (1.0714) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:41:43 loss 1.1341 (1.0880) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.290, TIME@all 0.312 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0628 (1.0748) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1630 (1.0929) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.319, TIME@all 0.312 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:41:54 loss 1.0920 (1.0747) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1240 (1.0889) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.422, TIME@all 0.312 +epoch: [190/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0547 (1.0809) acc 100.0000 (99.6875) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1035 (1.0904) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.368, TIME@all 0.312 +epoch: [190/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0931 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.0784 (1.0835) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.367, TIME@all 0.312 +epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:41:54 loss 1.0701 (1.0773) acc 100.0000 (99.8438) lr 0.026000 +epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:41:42 loss 1.0862 (1.0882) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.364, TIME@all 0.312 +epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:41:42 loss 1.0992 (1.0748) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:27 loss 1.1787 (1.0911) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.216, TIME@all 0.312 +epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1026 (1.0769) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1635 (1.0843) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.146, TIME@all 0.313 +epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.0933 (1.0764) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1343 (1.0833) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.111, TIME@all 0.313 +epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1142 (1.0794) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1196 (1.0903) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.084, TIME@all 0.313 +epoch: [191/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.0963 (1.0777) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:41:27 loss 1.1312 (1.0915) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.136, TIME@all 0.313 +epoch: [191/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:41:43 loss 1.0751 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:41:27 loss 1.2022 (1.0866) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 819.130, TIME@all 0.313 +epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1993 (1.0860) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1359 (1.0894) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.099, TIME@all 0.313 +epoch: [191/350][20/50] time 0.316 (0.314) data 0.001 (0.012) eta 0:41:43 loss 1.0862 (1.0737) acc 100.0000 (99.8438) lr 0.026000 +epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1705 (1.0892) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.150, TIME@all 0.313 +epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:24 loss 1.0751 (1.0709) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:41:14 loss 1.1225 (1.0899) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.540, TIME@all 0.312 +epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1043 (1.0691) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.2503 (1.0904) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.425, TIME@all 0.312 +epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0770 (1.0791) acc 100.0000 (99.6875) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.0817 (1.0858) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.445, TIME@all 0.312 +epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0724 (1.0683) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:14 loss 1.0960 (1.0848) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.462, TIME@all 0.312 +epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:41:25 loss 1.0802 (1.0737) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:41:15 loss 1.1527 (1.0931) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.442, TIME@all 0.312 +epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1003 (1.0717) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1241 (1.0822) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.478, TIME@all 0.312 +epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0635 (1.0733) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1435 (1.0925) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.490, TIME@all 0.312 +epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1038 (1.0712) acc 100.0000 (100.0000) lr 0.026000 +epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1047 (1.0832) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.476, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.0596 (1.0721) acc 100.0000 (99.8438) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.0818 (1.0817) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.336, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.0836 (1.0730) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.1505 (1.0855) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 820.245, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:01 loss 1.0662 (1.0786) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1196 (1.0865) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.167, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0692 (1.0698) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1867 (1.0818) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 820.222, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.1091 (1.0857) acc 100.0000 (99.8438) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.1121 (1.0880) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.238, TIME@all 0.312 +epoch: [193/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.1040 (1.0791) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.0852 (1.0796) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 820.226, TIME@all 0.312 +epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0816 (1.0791) acc 100.0000 (99.6875) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1292 (1.0857) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.234, TIME@all 0.312 +epoch: [193/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0940 (1.0797) acc 100.0000 (100.0000) lr 0.026000 +epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1539 (1.0844) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 820.225, TIME@all 0.312 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:58 loss 1.1249 (1.0852) acc 100.0000 (99.6875) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0630 (1.0917) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 814.875, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.1177 (1.0774) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0804 (1.0889) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.837, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.011) eta 0:40:59 loss 1.0782 (1.0758) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0755 (1.0817) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.765, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0740 (1.0711) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.1151 (1.0827) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.683, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.1320 (1.0824) acc 100.0000 (99.6875) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0933 (1.0859) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 814.782, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:58 loss 1.0941 (1.0737) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0646 (1.0896) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.864, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0655 (1.0712) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.1715 (1.0873) acc 96.8750 (99.9219) lr 0.026000 +FPS@all 814.779, TIME@all 0.314 +epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0687 (1.0744) acc 100.0000 (100.0000) lr 0.026000 +epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0723 (1.0818) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 814.804, TIME@all 0.314 +epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1256 (1.0767) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1168 (1.0888) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.086, TIME@all 0.313 +epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:35 loss 1.1625 (1.0807) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1465 (1.0937) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 817.931, TIME@all 0.313 +epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1195 (1.0751) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1115 (1.0882) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.017, TIME@all 0.313 +epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.0772 (1.0763) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1221 (1.0889) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 817.979, TIME@all 0.313 +epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.0884 (1.0734) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.0797 (1.0807) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 817.982, TIME@all 0.313 +epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1106 (1.0741) acc 100.0000 (100.0000) lr 0.026000 +epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:33 loss 1.1217 (1.0848) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.000, TIME@all 0.313 +epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:40:35 loss 1.0772 (1.0741) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:33 loss 1.0877 (1.0867) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 817.957, TIME@all 0.313 +epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1295 (1.0838) acc 100.0000 (99.8438) lr 0.026000 +epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1331 (1.0981) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 817.997, TIME@all 0.313 +epoch: [196/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.0739 (1.0758) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:40:06 loss 1.1075 (1.0827) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.194, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.0734 (1.0755) acc 100.0000 (99.8438) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 0:40:07 loss 1.1001 (1.0881) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 821.102, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.1167 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 0:40:07 loss 1.0582 (1.0881) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 821.140, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.0679 (1.0814) acc 100.0000 (99.6875) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.1522 (1.0879) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 821.072, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.1188 (1.0793) acc 100.0000 (99.8438) lr 0.026000 +epoch: [196/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:40:06 loss 1.1090 (1.0856) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 821.160, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.1483 (1.0859) acc 100.0000 (99.6875) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:40:07 loss 1.0743 (1.0942) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 821.085, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.1027 (1.0743) acc 100.0000 (100.0000) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.1133 (1.0876) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 821.096, TIME@all 0.312 +epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.0927 (1.0833) acc 100.0000 (99.6875) lr 0.026000 +epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.0887 (1.0881) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 821.106, TIME@all 0.312 +epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0733 (1.0738) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:08 loss 1.0862 (1.0890) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.732, TIME@all 0.314 +epoch: [197/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0572 (1.0715) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1123 (1.0845) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.643, TIME@all 0.314 +epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0634 (1.0794) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0982 (1.0911) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.596, TIME@all 0.314 +epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:16 loss 1.0723 (1.0728) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:08 loss 1.1479 (1.0880) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 814.756, TIME@all 0.314 +epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0657 (1.0689) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0859 (1.0956) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 814.607, TIME@all 0.314 +epoch: [197/350][20/50] time 0.319 (0.315) data 0.001 (0.013) eta 0:40:18 loss 1.0705 (1.0777) acc 100.0000 (100.0000) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1273 (1.0895) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.617, TIME@all 0.314 +epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0664 (1.0750) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1277 (1.0911) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 814.654, TIME@all 0.314 +epoch: [197/350][20/50] time 0.319 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0891 (1.0829) acc 100.0000 (99.8438) lr 0.026000 +epoch: [197/350][40/50] time 0.316 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0942 (1.0951) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 814.676, TIME@all 0.314 +epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.013) eta 0:42:08 loss 1.0687 (1.0764) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.315 (0.321) data 0.000 (0.007) eta 0:40:45 loss 1.0764 (1.0888) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 802.518, TIME@all 0.319 +epoch: [198/350][20/50] time 0.329 (0.331) data 0.001 (0.014) eta 0:42:06 loss 1.1264 (1.0802) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.315 (0.321) data 0.001 (0.007) eta 0:40:44 loss 1.0906 (1.0909) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 802.769, TIME@all 0.319 +epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.012) eta 0:42:09 loss 1.0881 (1.0721) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.006) eta 0:40:45 loss 1.0808 (1.0801) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 802.338, TIME@all 0.319 +epoch: [198/350][20/50] time 0.329 (0.331) data 0.001 (0.013) eta 0:42:06 loss 1.0723 (1.0688) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:44 loss 1.0827 (1.0882) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 802.577, TIME@all 0.319 +epoch: [198/350][20/50] time 0.328 (0.331) data 0.001 (0.013) eta 0:42:08 loss 1.1038 (1.0802) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:40:44 loss 1.0817 (1.0878) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 802.496, TIME@all 0.319 +epoch: [198/350][20/50] time 0.328 (0.331) data 0.000 (0.013) eta 0:42:08 loss 1.1007 (1.0734) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:40:45 loss 1.0859 (1.0841) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 802.371, TIME@all 0.319 +epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.013) eta 0:42:09 loss 1.0897 (1.0741) acc 100.0000 (100.0000) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:45 loss 1.0615 (1.0808) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 802.359, TIME@all 0.319 +epoch: [198/350][20/50] time 0.328 (0.331) data 0.001 (0.013) eta 0:42:06 loss 1.1104 (1.0751) acc 100.0000 (99.8438) lr 0.026000 +epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:44 loss 1.0714 (1.0821) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 802.599, TIME@all 0.319 +epoch: [199/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 0:39:27 loss 1.1758 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.308 (0.313) data 0.001 (0.007) eta 0:39:25 loss 1.1781 (1.0932) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 818.955, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:39:28 loss 1.2542 (1.0821) acc 96.8750 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:39:26 loss 1.2125 (1.0929) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 818.549, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:28 loss 1.2710 (1.0754) acc 96.8750 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.1396 (1.0970) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 818.503, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:29 loss 1.1252 (1.0711) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.0917 (1.0832) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 818.466, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:28 loss 1.1661 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.0994 (1.0861) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.552, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:29 loss 1.1515 (1.0774) acc 100.0000 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:39:25 loss 1.1166 (1.0886) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 818.766, TIME@all 0.313 +epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:39:28 loss 1.1334 (1.0713) acc 100.0000 (100.0000) lr 0.026000 +epoch: [199/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:39:25 loss 1.1403 (1.0881) acc 96.8750 (99.6094) lr 0.026000 +FPS@all 818.787, TIME@all 0.313 +epoch: [199/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:39:29 loss 1.2339 (1.0790) acc 96.8750 (99.8438) lr 0.026000 +epoch: [199/350][40/50] time 0.306 (0.313) data 0.000 (0.007) eta 0:39:26 loss 1.1948 (1.0869) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.517, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.013) eta 0:39:26 loss 1.1746 (1.0818) acc 100.0000 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:39:12 loss 1.1316 (1.0973) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.913, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0809 (1.0829) acc 100.0000 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:39:12 loss 1.1079 (1.0947) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 818.862, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0994 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:39:13 loss 1.0788 (1.0846) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.746, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0823 (1.0720) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:39:13 loss 1.1439 (1.0900) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.782, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0816 (1.0902) acc 100.0000 (99.5312) lr 0.026000 +epoch: [200/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:39:12 loss 1.1273 (1.0977) acc 96.8750 (99.5312) lr 0.026000 +FPS@all 818.788, TIME@all 0.313 +epoch: [200/350][20/50] time 0.319 (0.314) data 0.001 (0.013) eta 0:39:26 loss 1.1205 (1.0863) acc 100.0000 (99.8438) lr 0.026000 +epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:39:12 loss 1.1079 (1.0879) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.811, TIME@all 0.313 +epoch: [200/350][20/50] time 0.319 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.1487 (1.0749) acc 100.0000 (100.0000) lr 0.026000 +epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:39:12 loss 1.0620 (1.0884) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.819, TIME@all 0.313 +epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0940 (1.0810) acc 100.0000 (99.6875) lr 0.026000 +epoch: [200/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:39:12 loss 1.1006 (1.0904) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.782, TIME@all 0.313 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:38:58 loss 1.0926 (1.0804) acc 100.0000 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0820 (1.0916) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.634, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1669 (1.0810) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0640 (1.0887) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.462, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1348 (1.0839) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0718 (1.0940) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.560, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1551 (1.0756) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.001 (0.007) eta 0:38:53 loss 1.0869 (1.0886) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.504, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:38:59 loss 1.1544 (1.0780) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.001 (0.006) eta 0:38:53 loss 1.0662 (1.0893) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.460, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.0825 (1.0791) acc 100.0000 (99.6875) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0702 (1.0906) acc 100.0000 (99.4531) lr 0.026000 +FPS@all 819.462, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.0969 (1.0880) acc 100.0000 (99.8438) lr 0.026000 +epoch: [201/350][40/50] time 0.307 (0.313) data 0.001 (0.007) eta 0:38:53 loss 1.0702 (1.0922) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.504, TIME@all 0.312 +epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1848 (1.0840) acc 100.0000 (100.0000) lr 0.026000 +epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0799 (1.0953) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.523, TIME@all 0.312 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0996 (1.0764) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0822 (1.0810) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.592, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0738 (1.0700) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0837 (1.0766) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.441, TIME@all 0.313 +epoch: [202/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.0800 (1.0752) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0844 (1.0847) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.439, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0706 (1.0758) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0663 (1.0806) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.482, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.0885 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0638 (1.0834) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.500, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0674 (1.0783) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0771 (1.0834) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.539, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:38:41 loss 1.0720 (1.0761) acc 100.0000 (100.0000) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0655 (1.0852) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.492, TIME@all 0.313 +epoch: [202/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.1347 (1.0747) acc 100.0000 (99.8438) lr 0.026000 +epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0762 (1.0845) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 818.494, TIME@all 0.313 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1681 (1.0810) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:38:23 loss 1.1034 (1.0879) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.375, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.0832 (1.0664) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0880 (1.0784) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.301, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:31 loss 1.1336 (1.0852) acc 100.0000 (99.6875) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.1688 (1.0903) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.217, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:32 loss 1.1494 (1.0787) acc 96.8750 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.3765 (1.0931) acc 93.7500 (99.6875) lr 0.026000 +FPS@all 819.167, TIME@all 0.313 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1010 (1.0710) acc 100.0000 (100.0000) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 0:38:24 loss 1.0948 (1.0850) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.257, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1199 (1.0796) acc 100.0000 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:38:24 loss 1.1465 (1.0937) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.208, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:32 loss 1.0993 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0945 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.228, TIME@all 0.312 +epoch: [203/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:38:31 loss 1.1322 (1.0815) acc 96.8750 (99.6875) lr 0.026000 +epoch: [203/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0862 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.290, TIME@all 0.312 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0596 (1.0701) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0725 (1.0823) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.792, TIME@all 0.313 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0689 (1.0760) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0579 (1.0893) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.863, TIME@all 0.313 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:38:11 loss 1.0718 (1.0762) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0669 (1.0871) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.722, TIME@all 0.313 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0568 (1.0692) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0621 (1.0850) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.759, TIME@all 0.313 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0643 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +epoch: [204/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0878 (1.0891) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.690, TIME@all 0.313 +epoch: [204/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0752 (1.0710) acc 100.0000 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.001 (0.006) eta 0:38:08 loss 1.0666 (1.0848) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 818.754, TIME@all 0.313 +epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0637 (1.0790) acc 100.0000 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:38:09 loss 1.0547 (1.0843) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.755, TIME@all 0.313 +epoch: [204/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 0:38:12 loss 1.0769 (1.0744) acc 100.0000 (99.8438) lr 0.026000 +epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0548 (1.0898) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.735, TIME@all 0.313 +epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0637 (1.0697) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.1107 (1.0815) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 817.721, TIME@all 0.313 +epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0908 (1.0674) acc 100.0000 (99.8438) lr 0.026000 +epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:38:01 loss 1.0590 (1.0827) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 817.736, TIME@all 0.313 +epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0628 (1.0630) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.0866 (1.0762) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 817.797, TIME@all 0.313 +epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0860 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.313 (0.314) data 0.001 (0.006) eta 0:38:01 loss 1.0960 (1.0775) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 817.637, TIME@all 0.313 +epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0740 (1.0694) acc 100.0000 (99.8438) lr 0.026000 +epoch: [205/350][40/50] time 0.314 (0.314) data 0.001 (0.006) eta 0:38:01 loss 1.0916 (1.0778) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 817.692, TIME@all 0.313 +epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0966 (1.0683) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.0644 (1.0768) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 817.700, TIME@all 0.313 +epoch: [205/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.1020 (1.0743) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:38:01 loss 1.0602 (1.0777) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 817.639, TIME@all 0.313 +epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0993 (1.0723) acc 100.0000 (100.0000) lr 0.026000 +epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:38:01 loss 1.0654 (1.0801) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 817.644, TIME@all 0.313 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0661 (1.0722) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.0878 (1.0809) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 804.372, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:01 loss 1.0705 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0889 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 804.187, TIME@all 0.318 +epoch: [206/350][20/50] time 0.331 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.0610 (1.0628) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0747 (1.0739) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 804.327, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0743 (1.0762) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0822 (1.0831) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 804.264, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0706 (1.0731) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.1281 (1.0806) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 804.401, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:01 loss 1.0686 (1.0724) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.1103 (1.0847) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 804.311, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.0793 (1.0775) acc 100.0000 (99.8438) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.001 (0.006) eta 0:38:22 loss 1.0635 (1.0861) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 804.475, TIME@all 0.318 +epoch: [206/350][20/50] time 0.332 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.1042 (1.0681) acc 100.0000 (100.0000) lr 0.026000 +epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.0919 (1.0802) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 804.349, TIME@all 0.318 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.012) eta 0:37:41 loss 1.1307 (1.0747) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.006) eta 0:37:36 loss 1.0892 (1.0858) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.924, TIME@all 0.314 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.013) eta 0:37:42 loss 1.1756 (1.0825) acc 100.0000 (99.3750) lr 0.026000 +epoch: [207/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 0:37:37 loss 1.1178 (1.0969) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 814.627, TIME@all 0.314 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1394 (1.0773) acc 100.0000 (100.0000) lr 0.026000 +epoch: [207/350][40/50] time 0.325 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.0854 (1.0975) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 814.524, TIME@all 0.314 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1168 (1.0732) acc 100.0000 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.0980 (1.0856) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.484, TIME@all 0.314 +epoch: [207/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1201 (1.0670) acc 96.8750 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.327 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.1424 (1.0829) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 814.574, TIME@all 0.314 +epoch: [207/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:37:42 loss 1.1784 (1.0762) acc 96.8750 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.007) eta 0:37:37 loss 1.0983 (1.0831) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.529, TIME@all 0.314 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.012) eta 0:37:40 loss 1.1626 (1.0704) acc 96.8750 (99.8438) lr 0.026000 +epoch: [207/350][40/50] time 0.326 (0.315) data 0.001 (0.006) eta 0:37:36 loss 1.0691 (1.0939) acc 100.0000 (99.3750) lr 0.026000 +FPS@all 814.901, TIME@all 0.314 +epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.013) eta 0:37:40 loss 1.0863 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [207/350][40/50] time 0.326 (0.315) data 0.001 (0.007) eta 0:37:36 loss 1.1196 (1.0818) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 814.877, TIME@all 0.314 +epoch: [208/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0693 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0740 (1.0836) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.153, TIME@all 0.313 +epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0533 (1.0739) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0745 (1.0838) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.232, TIME@all 0.313 +epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:37:11 loss 1.0925 (1.0818) acc 100.0000 (99.6875) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:37:05 loss 1.0903 (1.0913) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 818.127, TIME@all 0.313 +epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0670 (1.0729) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0721 (1.0825) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.056, TIME@all 0.313 +epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0541 (1.0709) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0709 (1.0844) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.135, TIME@all 0.313 +epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.1007 (1.0768) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0749 (1.0866) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 818.118, TIME@all 0.313 +epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0897 (1.0727) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0657 (1.0844) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 818.086, TIME@all 0.313 +epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0683 (1.0724) acc 100.0000 (100.0000) lr 0.026000 +epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.1045 (1.0801) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 818.192, TIME@all 0.313 +epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.1159 (1.0750) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0766 (1.0844) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.639, TIME@all 0.314 +epoch: [209/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.0991 (1.0719) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:37:01 loss 1.0729 (1.0899) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 814.696, TIME@all 0.314 +epoch: [209/350][20/50] time 0.318 (0.315) data 0.001 (0.012) eta 0:37:11 loss 1.0702 (1.0711) acc 100.0000 (99.6875) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0860 (1.0899) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 814.569, TIME@all 0.314 +epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.011) eta 0:37:12 loss 1.1098 (1.0680) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0910 (1.0798) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 814.577, TIME@all 0.314 +epoch: [209/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 0:37:11 loss 1.1089 (1.0751) acc 96.8750 (99.5312) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.007) eta 0:37:02 loss 1.0600 (1.0857) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 814.641, TIME@all 0.314 +epoch: [209/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.1456 (1.0739) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.001 (0.006) eta 0:37:02 loss 1.0763 (1.0873) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 814.640, TIME@all 0.314 +epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:37:11 loss 1.0564 (1.0664) acc 100.0000 (99.8438) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.001 (0.007) eta 0:37:02 loss 1.0866 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 814.601, TIME@all 0.314 +epoch: [209/350][20/50] time 0.317 (0.315) data 0.001 (0.012) eta 0:37:11 loss 1.0866 (1.0698) acc 100.0000 (100.0000) lr 0.026000 +epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.1208 (1.0824) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 814.647, TIME@all 0.314 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0514 (1.0645) acc 100.0000 (99.8438) lr 0.026000 +epoch: [210/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0947 (1.0791) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.172, TIME@all 0.313 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0861 (1.0732) acc 100.0000 (99.8438) lr 0.026000 +epoch: [210/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0997 (1.0850) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.220, TIME@all 0.312 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0728 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0680 (1.0814) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.060, TIME@all 0.313 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0767 (1.0771) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0694 (1.0849) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.101, TIME@all 0.313 +epoch: [210/350][20/50] time 0.318 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0703 (1.0693) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0709 (1.0797) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.080, TIME@all 0.313 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0566 (1.0664) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0884 (1.0761) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.122, TIME@all 0.313 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:43 loss 1.0535 (1.0684) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:36:34 loss 1.0724 (1.0800) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.182, TIME@all 0.313 +epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0673 (1.0673) acc 100.0000 (100.0000) lr 0.026000 +epoch: [210/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:36:34 loss 1.1013 (1.0808) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.165, TIME@all 0.313 +epoch: [211/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:36:21 loss 1.0937 (1.0717) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0589 (1.0840) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.196, TIME@all 0.312 +epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.0611 (1.0712) acc 100.0000 (99.8438) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0634 (1.0801) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.097, TIME@all 0.312 +epoch: [211/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.1027 (1.0717) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0530 (1.0760) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.023, TIME@all 0.312 +epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:22 loss 1.0793 (1.0709) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0740 (1.0847) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.003, TIME@all 0.312 +epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.0805 (1.0717) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.1023 (1.0845) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.070, TIME@all 0.312 +epoch: [211/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:36:22 loss 1.0646 (1.0718) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0578 (1.0828) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.039, TIME@all 0.312 +epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:36:21 loss 1.0811 (1.0693) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0754 (1.0831) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.115, TIME@all 0.312 +epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:36:21 loss 1.0698 (1.0740) acc 100.0000 (100.0000) lr 0.026000 +epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0585 (1.0841) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.089, TIME@all 0.312 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0732 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1264 (1.0895) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.163, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:13 loss 1.1018 (1.0677) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.0856 (1.0868) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 816.112, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:13 loss 1.0760 (1.0743) acc 100.0000 (99.8438) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1284 (1.1000) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 815.989, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:13 loss 1.0769 (1.0730) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1292 (1.0893) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 816.027, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0523 (1.0649) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1274 (1.0824) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.097, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:12 loss 1.0563 (1.0686) acc 100.0000 (99.8438) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1493 (1.0835) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.115, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:12 loss 1.0625 (1.0695) acc 100.0000 (100.0000) lr 0.026000 +epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1716 (1.0962) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.031, TIME@all 0.314 +epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0726 (1.0749) acc 100.0000 (99.8438) lr 0.026000 +epoch: [212/350][40/50] time 0.323 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.2003 (1.0973) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 816.035, TIME@all 0.314 +epoch: [213/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:35:56 loss 1.1226 (1.0787) acc 100.0000 (99.8438) lr 0.026000 +epoch: [213/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.1145 (1.0847) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.148, TIME@all 0.312 +epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0834 (1.0704) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:35:45 loss 1.1028 (1.0861) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.166, TIME@all 0.312 +epoch: [213/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0854 (1.0709) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0706 (1.0782) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.039, TIME@all 0.312 +epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0746 (1.0666) acc 100.0000 (100.0000) lr 0.026000 +epoch: [213/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0644 (1.0812) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.047, TIME@all 0.312 +epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0529 (1.0794) acc 100.0000 (99.8438) lr 0.026000 +epoch: [213/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:35:45 loss 1.0677 (1.0861) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.088, TIME@all 0.312 +epoch: [213/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0753 (1.0880) acc 100.0000 (99.6875) lr 0.026000 +epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.1020 (1.0926) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 820.055, TIME@all 0.312 +epoch: [213/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:35:54 loss 1.1170 (1.0875) acc 100.0000 (99.6875) lr 0.026000 +epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0815 (1.0875) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 820.102, TIME@all 0.312 +epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0999 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:35:45 loss 1.0686 (1.0837) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 820.066, TIME@all 0.312 +epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1259 (1.0778) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1293 (1.0896) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 820.046, TIME@all 0.312 +epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:35:31 loss 1.0950 (1.0706) acc 100.0000 (99.8438) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1587 (1.0906) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.911, TIME@all 0.312 +epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1052 (1.0870) acc 100.0000 (99.6875) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:30 loss 1.0808 (1.0970) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.895, TIME@all 0.312 +epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.0807 (1.0739) acc 100.0000 (99.8438) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1208 (1.0931) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.924, TIME@all 0.312 +epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.2055 (1.0825) acc 96.8750 (99.6875) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1016 (1.0968) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.953, TIME@all 0.312 +epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1763 (1.0815) acc 100.0000 (100.0000) lr 0.026000 +epoch: [214/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1137 (1.1012) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.896, TIME@all 0.312 +epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.0656 (1.0763) acc 100.0000 (99.8438) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1402 (1.0981) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.956, TIME@all 0.312 +epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1417 (1.0795) acc 100.0000 (99.6875) lr 0.026000 +epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.0746 (1.0882) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.978, TIME@all 0.312 +epoch: [215/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:35:25 loss 1.0658 (1.0776) acc 100.0000 (99.8438) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:35:23 loss 1.1430 (1.0855) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.557, TIME@all 0.314 +epoch: [215/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:35:26 loss 1.1074 (1.0759) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0877 (1.0863) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.413, TIME@all 0.314 +epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1706 (1.0853) acc 96.8750 (99.6875) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0865 (1.0865) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.357, TIME@all 0.314 +epoch: [215/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1601 (1.0801) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0905 (1.0851) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 816.412, TIME@all 0.314 +epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.0970 (1.0758) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0928 (1.0905) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 816.467, TIME@all 0.314 +epoch: [215/350][20/50] time 0.317 (0.314) data 0.001 (0.013) eta 0:35:25 loss 1.0839 (1.0745) acc 100.0000 (99.6875) lr 0.026000 +epoch: [215/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:35:24 loss 1.0830 (1.0861) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.422, TIME@all 0.314 +epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1414 (1.0796) acc 100.0000 (100.0000) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:23 loss 1.0730 (1.0879) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.480, TIME@all 0.314 +epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.0676 (1.0741) acc 100.0000 (99.6875) lr 0.026000 +epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:23 loss 1.0796 (1.0855) acc 100.0000 (99.5312) lr 0.026000 +FPS@all 816.470, TIME@all 0.314 +epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:35:10 loss 1.0997 (1.0650) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:59 loss 1.0779 (1.0899) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.378, TIME@all 0.312 +epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1235 (1.0687) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.1766 (1.0841) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.252, TIME@all 0.312 +epoch: [216/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1103 (1.0746) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.1160 (1.0850) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.274, TIME@all 0.312 +epoch: [216/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:35:11 loss 1.1132 (1.0701) acc 100.0000 (99.8438) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:59 loss 1.1453 (1.0894) acc 96.8750 (99.4531) lr 0.026000 +FPS@all 819.342, TIME@all 0.312 +epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:35:11 loss 1.0650 (1.0742) acc 100.0000 (99.6875) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:34:59 loss 1.0754 (1.0852) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.327, TIME@all 0.312 +epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1218 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:34:59 loss 1.1002 (1.0850) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.309, TIME@all 0.312 +epoch: [216/350][20/50] time 0.311 (0.314) data 0.001 (0.012) eta 0:35:11 loss 1.1216 (1.0748) acc 100.0000 (99.8438) lr 0.026000 +epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.0931 (1.0824) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.317, TIME@all 0.312 +epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1444 (1.0760) acc 100.0000 (100.0000) lr 0.026000 +epoch: [216/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:34:59 loss 1.0942 (1.0943) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.336, TIME@all 0.312 +epoch: [217/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:34:47 loss 1.1087 (1.0742) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:41 loss 1.1219 (1.0859) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.505, TIME@all 0.312 +epoch: [217/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.1404 (1.0808) acc 100.0000 (99.6875) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.0999 (1.0884) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.340, TIME@all 0.312 +epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1746 (1.0717) acc 96.8750 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:42 loss 1.1367 (1.0832) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.317, TIME@all 0.312 +epoch: [217/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:34:47 loss 1.0993 (1.0750) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.1665 (1.0836) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.418, TIME@all 0.312 +epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1177 (1.0782) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:34:42 loss 1.2443 (1.0918) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.410, TIME@all 0.312 +epoch: [217/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.1002 (1.0733) acc 100.0000 (100.0000) lr 0.026000 +epoch: [217/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.1234 (1.0852) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.388, TIME@all 0.312 +epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1041 (1.0736) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:42 loss 1.2797 (1.0926) acc 93.7500 (99.6094) lr 0.026000 +FPS@all 819.357, TIME@all 0.312 +epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.0811 (1.0789) acc 100.0000 (99.8438) lr 0.026000 +epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.0862 (1.0931) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.326, TIME@all 0.312 +epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0983 (1.0684) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0568 (1.0825) acc 100.0000 (99.6875) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.199, TIME@all 0.312 +epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:34:33 loss 1.0764 (1.0724) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.1151 (1.0873) acc 100.0000 (99.8438) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.303, TIME@all 0.312 +epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:34:34 loss 1.0659 (1.0747) acc 100.0000 (99.8438) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:29 loss 1.0795 (1.0859) acc 100.0000 (99.6875) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.126, TIME@all 0.312 +epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.1033 (1.0710) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:29 loss 1.0869 (1.0772) acc 100.0000 (99.8438) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.096, TIME@all 0.312 +epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0567 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:29 loss 1.0736 (1.0862) acc 100.0000 (100.0000) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.180, TIME@all 0.312 +epoch: [218/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 0:34:33 loss 1.0703 (1.0750) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0803 (1.0846) acc 100.0000 (99.8438) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.244, TIME@all 0.312 +epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:33 loss 1.0747 (1.0689) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0561 (1.0762) acc 100.0000 (99.9219) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.174, TIME@all 0.312 +epoch: [218/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0535 (1.0670) acc 100.0000 (100.0000) lr 0.026000 +epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.1097 (1.0888) acc 100.0000 (99.7656) lr 0.026000 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 +FPS@all 820.200, TIME@all 0.312 +epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:18 loss 1.0585 (1.0712) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:12 loss 1.0922 (1.0899) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.685, TIME@all 0.312 +epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0615 (1.0667) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:34:13 loss 1.0571 (1.0777) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.530, TIME@all 0.312 +epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:18 loss 1.0594 (1.0661) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0776 (1.0834) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.621, TIME@all 0.312 +epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0775 (1.0691) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0570 (1.0848) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 819.492, TIME@all 0.312 +epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0661 (1.0775) acc 100.0000 (99.6875) lr 0.026000 +epoch: [219/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:34:13 loss 1.0620 (1.0886) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.516, TIME@all 0.312 +epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0550 (1.0696) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:13 loss 1.0698 (1.0827) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.606, TIME@all 0.312 +epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0714 (1.0732) acc 100.0000 (99.8438) lr 0.026000 +epoch: [219/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:34:13 loss 1.1110 (1.0904) acc 100.0000 (99.6094) lr 0.026000 +FPS@all 819.557, TIME@all 0.312 +epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0643 (1.0680) acc 100.0000 (100.0000) lr 0.026000 +epoch: [219/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0570 (1.0890) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.546, TIME@all 0.312 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:01 loss 1.0582 (1.0694) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:33:59 loss 1.0712 (1.0795) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 817.511, TIME@all 0.313 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.0837 (1.0672) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0869 (1.0769) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 817.424, TIME@all 0.313 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:34:02 loss 1.0750 (1.0688) acc 100.0000 (99.8438) lr 0.026000 +epoch: [220/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0812 (1.0818) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.336, TIME@all 0.313 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.1204 (1.0736) acc 96.8750 (99.8438) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0927 (1.0847) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 817.386, TIME@all 0.313 +epoch: [220/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:34:02 loss 1.0862 (1.0752) acc 100.0000 (99.6875) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:00 loss 1.0966 (1.0847) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.344, TIME@all 0.313 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.0904 (1.0707) acc 100.0000 (99.8438) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.1971 (1.0869) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 817.408, TIME@all 0.313 +epoch: [220/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:34:02 loss 1.0768 (1.0736) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:00 loss 1.0893 (1.0833) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 817.461, TIME@all 0.313 +epoch: [220/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:34:02 loss 1.0990 (1.0712) acc 100.0000 (100.0000) lr 0.026000 +epoch: [220/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:34:00 loss 1.0954 (1.0817) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 817.446, TIME@all 0.313 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:33:42 loss 1.1429 (1.0722) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:33:36 loss 1.1264 (1.0795) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.188, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1087 (1.0767) acc 100.0000 (99.8438) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0816 (1.0770) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.098, TIME@all 0.312 +epoch: [221/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1168 (1.0750) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.1172 (1.0839) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 821.025, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:43 loss 1.1077 (1.0813) acc 100.0000 (99.6875) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0750 (1.0809) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 820.972, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1582 (1.0761) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.1564 (1.0820) acc 96.8750 (99.8438) lr 0.026000 +FPS@all 821.049, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:33:43 loss 1.1509 (1.0875) acc 100.0000 (99.6875) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:33:36 loss 1.0745 (1.0810) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 821.068, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:43 loss 1.1519 (1.0763) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0887 (1.0801) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 821.107, TIME@all 0.312 +epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.0606 (1.0716) acc 100.0000 (100.0000) lr 0.026000 +epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0957 (1.0771) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 821.092, TIME@all 0.312 +epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0704 (1.0706) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1691 (1.0840) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.776, TIME@all 0.312 +epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0653 (1.0679) acc 100.0000 (99.8438) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0622 (1.0749) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.677, TIME@all 0.312 +epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0562 (1.0676) acc 100.0000 (99.8438) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0748 (1.0762) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.859, TIME@all 0.312 +epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0943 (1.0719) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1034 (1.0826) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.725, TIME@all 0.312 +epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0714 (1.0720) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0586 (1.0834) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.715, TIME@all 0.312 +epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0979 (1.0682) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0700 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.676, TIME@all 0.312 +epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0584 (1.0700) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1096 (1.0839) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.713, TIME@all 0.312 +epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:33:31 loss 1.0903 (1.0710) acc 100.0000 (100.0000) lr 0.026000 +epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0625 (1.0768) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 819.714, TIME@all 0.312 +epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:33:22 loss 1.0601 (1.0672) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:33:18 loss 1.0710 (1.0757) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.411, TIME@all 0.314 +epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0949 (1.0685) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1656 (1.0861) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.211, TIME@all 0.314 +epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0915 (1.0675) acc 96.8750 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1161 (1.0791) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 816.318, TIME@all 0.314 +epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0590 (1.0662) acc 100.0000 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.0918 (1.0824) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 816.314, TIME@all 0.314 +epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.1076 (1.0733) acc 100.0000 (99.8438) lr 0.026000 +epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1319 (1.0840) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 816.260, TIME@all 0.314 +epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0790 (1.0666) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1513 (1.0858) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 816.322, TIME@all 0.314 +epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:33:22 loss 1.0642 (1.0639) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:33:19 loss 1.1176 (1.0825) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 816.275, TIME@all 0.314 +epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0725 (1.0676) acc 100.0000 (100.0000) lr 0.026000 +epoch: [223/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1335 (1.0806) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 816.314, TIME@all 0.314 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0632 (1.0634) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1048 (1.0781) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.848, TIME@all 0.312 +epoch: [224/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0695 (1.0656) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.0905 (1.0817) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.698, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0706 (1.0688) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1221 (1.0792) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 819.742, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0814 (1.0770) acc 100.0000 (99.6875) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1817 (1.0850) acc 96.8750 (99.6875) lr 0.026000 +FPS@all 819.728, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0972 (1.0725) acc 100.0000 (100.0000) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1449 (1.0843) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.717, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0755 (1.0741) acc 100.0000 (99.8438) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:32:54 loss 1.1502 (1.0899) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 819.683, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.1541 (1.0775) acc 96.8750 (99.6875) lr 0.026000 +epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1221 (1.0813) acc 96.8750 (99.7656) lr 0.026000 +FPS@all 819.696, TIME@all 0.312 +epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0697 (1.0689) acc 100.0000 (99.8438) lr 0.026000 +epoch: [224/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.0827 (1.0818) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 819.675, TIME@all 0.312 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.1334 (1.0660) acc 96.8750 (99.8438) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.1423 (1.0792) acc 100.0000 (99.9219) lr 0.026000 +FPS@all 816.043, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0554 (1.0740) acc 100.0000 (99.8438) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.1159 (1.0868) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 815.903, TIME@all 0.314 +epoch: [225/350][20/50] time 0.315 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0554 (1.0681) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0967 (1.0874) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 815.904, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.1417 (1.0702) acc 96.8750 (99.8438) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.0725 (1.0826) acc 100.0000 (99.6875) lr 0.026000 +FPS@all 815.888, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0617 (1.0651) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0895 (1.0882) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 815.968, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.0692 (1.0631) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.0752 (1.0812) acc 100.0000 (99.7656) lr 0.026000 +FPS@all 815.926, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.1022 (1.0635) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0824 (1.0735) acc 100.0000 (100.0000) lr 0.026000 +FPS@all 815.928, TIME@all 0.314 +epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.0572 (1.0635) acc 100.0000 (100.0000) lr 0.026000 +epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0584 (1.0789) acc 100.0000 (99.8438) lr 0.026000 +FPS@all 815.959, TIME@all 0.314 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0864 (1.0690) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.1441 (1.0789) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.707, TIME@all 0.313 +epoch: [226/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.1636 (1.0702) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0812 (1.0806) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.663, TIME@all 0.313 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:30 loss 1.1490 (1.0737) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:32:25 loss 1.0615 (1.0763) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.614, TIME@all 0.313 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.2220 (1.0693) acc 96.8750 (99.8438) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:25 loss 1.0558 (1.0787) acc 100.0000 (99.5312) lr 0.002600 +FPS@all 818.604, TIME@all 0.313 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0870 (1.0690) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.1048 (1.0803) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.679, TIME@all 0.313 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0989 (1.0696) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0508 (1.0791) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.723, TIME@all 0.313 +epoch: [226/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0868 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:25 loss 1.0673 (1.0764) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.668, TIME@all 0.313 +epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.1114 (1.0714) acc 100.0000 (100.0000) lr 0.002600 +epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0863 (1.0809) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.710, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0913 (1.0681) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0627 (1.0703) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.440, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:32:18 loss 1.0829 (1.0687) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0601 (1.0731) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.503, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.1233 (1.0631) acc 96.8750 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0538 (1.0716) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.376, TIME@all 0.313 +epoch: [227/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:32:18 loss 1.0578 (1.0668) acc 100.0000 (99.6875) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:32:10 loss 1.0524 (1.0750) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 818.403, TIME@all 0.313 +epoch: [227/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0761 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0615 (1.0710) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.392, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0612 (1.0641) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0816 (1.0677) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.397, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0829 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:32:10 loss 1.0592 (1.0726) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.418, TIME@all 0.313 +epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0540 (1.0716) acc 100.0000 (99.6875) lr 0.002600 +epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0742 (1.0745) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.407, TIME@all 0.313 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:32:05 loss 1.0690 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0824 (1.0701) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 814.315, TIME@all 0.314 +epoch: [228/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0530 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0827 (1.0751) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 814.128, TIME@all 0.314 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:32:06 loss 1.0547 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.311 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0772 (1.0683) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 814.203, TIME@all 0.314 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0602 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0706 (1.0677) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.200, TIME@all 0.314 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0805 (1.0679) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.1366 (1.0761) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 814.140, TIME@all 0.314 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0871 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0771 (1.0763) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 814.231, TIME@all 0.314 +epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0745 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.001 (0.007) eta 0:32:08 loss 1.0914 (1.0717) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 814.212, TIME@all 0.314 +epoch: [228/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0519 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0503 (1.0687) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.160, TIME@all 0.314 +epoch: [229/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:31:45 loss 1.0635 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0540 (1.0724) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.769, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:31:44 loss 1.1461 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0710 (1.0702) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.800, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.0853 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:36 loss 1.0551 (1.0746) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.726, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.0848 (1.0595) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:36 loss 1.1294 (1.0672) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.677, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.1768 (1.0679) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0591 (1.0740) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.725, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:31:44 loss 1.1334 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0552 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.726, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:31:44 loss 1.1128 (1.0563) acc 96.8750 (99.8438) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0613 (1.0661) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.769, TIME@all 0.312 +epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.1244 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0485 (1.0727) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.718, TIME@all 0.312 +epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 0:31:23 loss 1.1052 (1.0669) acc 100.0000 (99.8438) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:31:20 loss 1.0944 (1.0769) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.414, TIME@all 0.312 +epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1507 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0668 (1.0749) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.147, TIME@all 0.313 +epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:31:24 loss 1.1880 (1.0819) acc 96.8750 (99.2188) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0757 (1.0858) acc 100.0000 (99.3750) lr 0.002600 +FPS@all 819.174, TIME@all 0.313 +epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:31:24 loss 1.1013 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.1022 (1.0750) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.249, TIME@all 0.312 +epoch: [230/350][20/50] time 0.316 (0.312) data 0.001 (0.012) eta 0:31:24 loss 1.0915 (1.0728) acc 100.0000 (99.8438) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0980 (1.0800) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.237, TIME@all 0.312 +epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1427 (1.0689) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:31:20 loss 1.1011 (1.0762) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.173, TIME@all 0.313 +epoch: [230/350][20/50] time 0.316 (0.313) data 0.001 (0.012) eta 0:31:24 loss 1.1038 (1.0717) acc 100.0000 (99.6875) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0724 (1.0797) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.188, TIME@all 0.313 +epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1286 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0859 (1.0741) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.189, TIME@all 0.313 +epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0566 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0510 (1.0744) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.767, TIME@all 0.313 +epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.014) eta 0:31:23 loss 1.0552 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0903 (1.0735) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.849, TIME@all 0.313 +epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.012) eta 0:31:23 loss 1.0529 (1.0651) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:31:12 loss 1.0590 (1.0737) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.733, TIME@all 0.313 +epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0534 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0935 (1.0768) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.699, TIME@all 0.313 +epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0885 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0492 (1.0717) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.743, TIME@all 0.313 +epoch: [231/350][20/50] time 0.327 (0.315) data 0.001 (0.013) eta 0:31:23 loss 1.0542 (1.0609) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0504 (1.0743) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.724, TIME@all 0.313 +epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0637 (1.0650) acc 100.0000 (99.8438) lr 0.002600 +epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0489 (1.0710) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.753, TIME@all 0.313 +epoch: [231/350][20/50] time 0.326 (0.315) data 0.001 (0.013) eta 0:31:23 loss 1.0529 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0735 (1.0767) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.786, TIME@all 0.313 +epoch: [232/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0776 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0637 (1.0645) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.869, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:30:47 loss 1.0784 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:44 loss 1.0504 (1.0656) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.966, TIME@all 0.312 +epoch: [232/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 0:30:47 loss 1.0960 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0599 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.815, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:30:47 loss 1.0588 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:30:45 loss 1.0643 (1.0649) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.800, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0842 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.1056 (1.0657) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 820.851, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0628 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0697 (1.0657) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.804, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:30:47 loss 1.0752 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0793 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.883, TIME@all 0.312 +epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0643 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [232/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0807 (1.0615) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.819, TIME@all 0.312 +epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:30:43 loss 1.0645 (1.0657) acc 100.0000 (99.8438) lr 0.002600 +epoch: [233/350][40/50] time 0.320 (0.314) data 0.000 (0.007) eta 0:30:40 loss 1.0629 (1.0768) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.110, TIME@all 0.313 +epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:30:43 loss 1.0599 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.007) eta 0:30:40 loss 1.0502 (1.0735) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.190, TIME@all 0.313 +epoch: [233/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:30:43 loss 1.0813 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0581 (1.0727) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.080, TIME@all 0.313 +epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0632 (1.0696) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0603 (1.0758) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.046, TIME@all 0.313 +epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0710 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0588 (1.0700) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.072, TIME@all 0.313 +epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0565 (1.0611) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0515 (1.0762) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.120, TIME@all 0.313 +epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0899 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0572 (1.0722) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.116, TIME@all 0.313 +epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0645 (1.0637) acc 100.0000 (99.8438) lr 0.002600 +epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0772 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.068, TIME@all 0.313 +epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0614 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0527 (1.0633) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.062, TIME@all 0.313 +epoch: [234/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:30:25 loss 1.0554 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0502 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.011, TIME@all 0.313 +epoch: [234/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:30:25 loss 1.0573 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:30:19 loss 1.0664 (1.0668) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.865, TIME@all 0.313 +epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0608 (1.0625) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0593 (1.0714) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.877, TIME@all 0.313 +epoch: [234/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:30:25 loss 1.0682 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0550 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.917, TIME@all 0.313 +epoch: [234/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0657 (1.0617) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0735 (1.0713) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.903, TIME@all 0.313 +epoch: [234/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:30:25 loss 1.0578 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0575 (1.0711) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.954, TIME@all 0.313 +epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0528 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0512 (1.0665) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.919, TIME@all 0.313 +epoch: [235/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.1035 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0590 (1.0637) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.936, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.0993 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.0794 (1.0698) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.973, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:09 loss 1.0429 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0715 (1.0668) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.815, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:09 loss 1.0643 (1.0546) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0560 (1.0614) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.784, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.0890 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.1042 (1.0640) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.901, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:08 loss 1.0774 (1.0571) acc 100.0000 (99.8438) lr 0.002600 +epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0714 (1.0697) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.870, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:09 loss 1.1745 (1.0635) acc 100.0000 (99.8438) lr 0.002600 +epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.1166 (1.0726) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.858, TIME@all 0.312 +epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:08 loss 1.0870 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0691 (1.0668) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.866, TIME@all 0.312 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:57 loss 1.1160 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:29:51 loss 1.0574 (1.0717) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.991, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0989 (1.0606) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0562 (1.0762) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.902, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0621 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:29:52 loss 1.0661 (1.0697) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.839, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:29:58 loss 1.0691 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:29:52 loss 1.0598 (1.0701) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.800, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0860 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0527 (1.0712) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.846, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0816 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.1109 (1.0727) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.836, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0620 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0488 (1.0738) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.845, TIME@all 0.313 +epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0695 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +epoch: [236/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:29:51 loss 1.0720 (1.0747) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.912, TIME@all 0.313 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.015) eta 0:29:35 loss 1.1694 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.314 (0.312) data 0.001 (0.008) eta 0:29:27 loss 1.1569 (1.0776) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.399, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.1039 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0566 (1.0696) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.366, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:29:35 loss 1.0868 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.1290 (1.0726) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 821.200, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:29:35 loss 1.1174 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0885 (1.0746) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.281, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:29:35 loss 1.1267 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0752 (1.0757) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 821.342, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.1203 (1.0690) acc 96.8750 (99.6875) lr 0.002600 +epoch: [237/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:29:27 loss 1.0887 (1.0736) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.345, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.0702 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.1399 (1.0767) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.257, TIME@all 0.312 +epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.0950 (1.0694) acc 100.0000 (100.0000) lr 0.002600 +epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0656 (1.0780) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.283, TIME@all 0.312 +epoch: [238/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:29:17 loss 1.0765 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:29:13 loss 1.0519 (1.0670) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.992, TIME@all 0.312 +epoch: [238/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0543 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0553 (1.0717) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.788, TIME@all 0.312 +epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0558 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0499 (1.0689) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.828, TIME@all 0.312 +epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0754 (1.0611) acc 100.0000 (99.8438) lr 0.002600 +epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0608 (1.0683) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.917, TIME@all 0.312 +epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0869 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0555 (1.0763) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.857, TIME@all 0.312 +epoch: [238/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0487 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0513 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.816, TIME@all 0.312 +epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0510 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0613 (1.0707) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.842, TIME@all 0.312 +epoch: [238/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0607 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0833 (1.0693) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.906, TIME@all 0.312 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0579 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0471 (1.0706) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.694, TIME@all 0.313 +epoch: [239/350][20/50] time 0.315 (0.315) data 0.000 (0.013) eta 0:29:16 loss 1.0539 (1.0674) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:29:04 loss 1.0590 (1.0795) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.769, TIME@all 0.313 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0653 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0580 (1.0720) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.578, TIME@all 0.313 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0539 (1.0619) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0658 (1.0708) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.518, TIME@all 0.313 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:29:17 loss 1.0530 (1.0590) acc 100.0000 (99.8438) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0496 (1.0683) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.616, TIME@all 0.313 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0620 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0505 (1.0704) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.590, TIME@all 0.313 +epoch: [239/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0550 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0551 (1.0714) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.616, TIME@all 0.313 +epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0529 (1.0615) acc 100.0000 (100.0000) lr 0.002600 +epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0547 (1.0760) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.614, TIME@all 0.313 +epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.1026 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.0943 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.265, TIME@all 0.312 +epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.1136 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:28:44 loss 1.0621 (1.0685) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.291, TIME@all 0.312 +epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0652 (1.0641) acc 100.0000 (99.8438) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.2136 (1.0712) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 819.181, TIME@all 0.313 +epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:28:53 loss 1.0811 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:28:45 loss 1.0928 (1.0699) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.178, TIME@all 0.313 +epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0519 (1.0640) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.1313 (1.0695) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.195, TIME@all 0.313 +epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.0605 (1.0611) acc 100.0000 (100.0000) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:44 loss 1.0693 (1.0686) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.261, TIME@all 0.312 +epoch: [240/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:28:53 loss 1.0798 (1.0690) acc 100.0000 (99.6875) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.0730 (1.0694) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.152, TIME@all 0.313 +epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0657 (1.0689) acc 100.0000 (99.6875) lr 0.002600 +epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.1144 (1.0706) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.211, TIME@all 0.312 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0560 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0532 (1.0677) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.518, TIME@all 0.313 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0636 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0563 (1.0633) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.547, TIME@all 0.313 +epoch: [241/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.0608 (1.0625) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.2085 (1.0735) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 817.423, TIME@all 0.313 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.1087 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.0537 (1.0749) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.401, TIME@all 0.313 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:33 loss 1.0602 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0714 (1.0640) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.421, TIME@all 0.313 +epoch: [241/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.0777 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.0624 (1.0695) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.494, TIME@all 0.313 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0824 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0744 (1.0726) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.427, TIME@all 0.313 +epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:33 loss 1.0559 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0750 (1.0680) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.474, TIME@all 0.313 +epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0699 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.314 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.0833 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 813.976, TIME@all 0.315 +epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.012) eta 0:28:35 loss 1.0623 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:27 loss 1.0706 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 814.006, TIME@all 0.314 +epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0772 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.0947 (1.0673) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 813.983, TIME@all 0.315 +epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0771 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:28 loss 1.1099 (1.0679) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 813.888, TIME@all 0.315 +epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.012) eta 0:28:35 loss 1.0609 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:28 loss 1.1502 (1.0672) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 813.853, TIME@all 0.315 +epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0670 (1.0655) acc 100.0000 (99.8438) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:28 loss 1.0785 (1.0722) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 813.889, TIME@all 0.315 +epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0673 (1.0642) acc 100.0000 (99.6875) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.1062 (1.0642) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 813.904, TIME@all 0.315 +epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0585 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:28 loss 1.0882 (1.0644) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 813.883, TIME@all 0.315 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0626 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0701 (1.0686) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 807.035, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0777 (1.0584) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0675 (1.0661) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 807.078, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.0645 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0590 (1.0707) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 806.922, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.001 (0.012) eta 0:28:42 loss 1.0594 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:27 loss 1.0640 (1.0704) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 807.174, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.0854 (1.0617) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.320 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0543 (1.0684) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 807.028, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0540 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:28 loss 1.0601 (1.0702) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 806.952, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.1466 (1.0625) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0606 (1.0721) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 806.961, TIME@all 0.317 +epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0702 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0472 (1.0702) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 807.137, TIME@all 0.317 +epoch: [244/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:27:48 loss 1.0561 (1.0573) acc 100.0000 (99.8438) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0536 (1.0639) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.691, TIME@all 0.314 +epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0669 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0855 (1.0623) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.715, TIME@all 0.314 +epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0922 (1.0624) acc 100.0000 (99.8438) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0625 (1.0727) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 815.563, TIME@all 0.314 +epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0681 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0970 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.612, TIME@all 0.314 +epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:27:48 loss 1.0542 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0568 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.550, TIME@all 0.314 +epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0853 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0741 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.705, TIME@all 0.314 +epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0495 (1.0530) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.319 (0.314) data 0.001 (0.006) eta 0:27:49 loss 1.0841 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.632, TIME@all 0.314 +epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0485 (1.0506) acc 100.0000 (100.0000) lr 0.002600 +epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0595 (1.0617) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 815.582, TIME@all 0.314 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:27:31 loss 1.0503 (1.0653) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0629 (1.0725) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.288, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0492 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0972 (1.0719) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.145, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:27:31 loss 1.0704 (1.0642) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0596 (1.0674) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.223, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:31 loss 1.0463 (1.0684) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0643 (1.0694) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.215, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:31 loss 1.0567 (1.0732) acc 100.0000 (99.8438) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.1027 (1.0768) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.148, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:27:31 loss 1.0567 (1.0648) acc 100.0000 (100.0000) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0616 (1.0714) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.160, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0618 (1.0613) acc 100.0000 (99.8438) lr 0.002600 +epoch: [245/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0568 (1.0709) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.106, TIME@all 0.312 +epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0577 (1.0676) acc 100.0000 (99.8438) lr 0.002600 +epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0608 (1.0712) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.135, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0810 (1.0740) acc 100.0000 (99.5312) lr 0.002600 +epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0674 (1.0736) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.936, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:15 loss 1.0475 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:27:11 loss 1.0635 (1.0677) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.900, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0716 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.1009 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.884, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0559 (1.0605) acc 100.0000 (99.8438) lr 0.002600 +epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0537 (1.0748) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.949, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0622 (1.0629) acc 100.0000 (99.8438) lr 0.002600 +epoch: [246/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:27:10 loss 1.0773 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.910, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0513 (1.0661) acc 100.0000 (99.8438) lr 0.002600 +epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0723 (1.0700) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.913, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0581 (1.0619) acc 100.0000 (99.8438) lr 0.002600 +epoch: [246/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0792 (1.0687) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.917, TIME@all 0.313 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0886 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +epoch: [246/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0917 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.889, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.0590 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0596 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.556, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:27:08 loss 1.0849 (1.0628) acc 96.8750 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:26:58 loss 1.0649 (1.0737) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.598, TIME@all 0.313 +epoch: [247/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.0834 (1.0593) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0798 (1.0667) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.508, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.011) eta 0:27:09 loss 1.0760 (1.0623) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:59 loss 1.0953 (1.0684) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.467, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.1665 (1.0702) acc 100.0000 (100.0000) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0782 (1.0698) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.533, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:27:09 loss 1.0901 (1.0640) acc 100.0000 (99.8438) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:26:58 loss 1.1353 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.534, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.1310 (1.0664) acc 96.8750 (99.6875) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0661 (1.0702) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.543, TIME@all 0.313 +epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:27:09 loss 1.2356 (1.0668) acc 96.8750 (99.6875) lr 0.002600 +epoch: [247/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:26:58 loss 1.0839 (1.0731) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 817.535, TIME@all 0.313 +epoch: [248/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0495 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0495 (1.0708) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 819.149, TIME@all 0.313 +epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0495 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0512 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.027, TIME@all 0.313 +epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0695 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0782 (1.0633) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.073, TIME@all 0.313 +epoch: [248/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0532 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0547 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.985, TIME@all 0.313 +epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0482 (1.0589) acc 100.0000 (99.8438) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0665 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.005, TIME@all 0.313 +epoch: [248/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:26:41 loss 1.0566 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0585 (1.0709) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 819.023, TIME@all 0.313 +epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0491 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0900 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.095, TIME@all 0.313 +epoch: [248/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0545 (1.0577) acc 100.0000 (100.0000) lr 0.002600 +epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0503 (1.0625) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.029, TIME@all 0.313 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:39 loss 1.1258 (1.0619) acc 96.8750 (99.8438) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0811 (1.0710) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 815.525, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:39 loss 1.0573 (1.0642) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0617 (1.0724) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 815.577, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.1247 (1.0653) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0547 (1.0661) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.401, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.012) eta 0:26:40 loss 1.0992 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0611 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.445, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.0710 (1.0602) acc 100.0000 (99.8438) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0521 (1.0699) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 815.489, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.0792 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0887 (1.0708) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 815.443, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.1630 (1.0690) acc 96.8750 (99.8438) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.001 (0.007) eta 0:26:30 loss 1.0570 (1.0761) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.471, TIME@all 0.314 +epoch: [249/350][20/50] time 0.328 (0.315) data 0.001 (0.013) eta 0:26:39 loss 1.0750 (1.0621) acc 100.0000 (99.8438) lr 0.002600 +epoch: [249/350][40/50] time 0.318 (0.314) data 0.001 (0.006) eta 0:26:30 loss 1.0745 (1.0685) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.524, TIME@all 0.314 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.001 (0.013) eta 0:26:21 loss 1.1160 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:10 loss 1.0966 (1.0716) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.619, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0671 (1.0640) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0934 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.535, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0676 (1.0556) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:10 loss 1.0834 (1.0700) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.442, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0582 (1.0651) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0705 (1.0747) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 818.466, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0783 (1.0671) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0702 (1.0777) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.497, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:26:21 loss 1.0735 (1.0608) acc 100.0000 (99.8438) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0945 (1.0695) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.509, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0511 (1.0665) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.1106 (1.0732) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.532, TIME@all 0.313 +epoch: [250/350][20/50] time 0.309 (0.314) data 0.001 (0.012) eta 0:26:21 loss 1.0705 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0628 (1.0678) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.463, TIME@all 0.313 +epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.0852 (1.0608) acc 100.0000 (99.6875) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.001 (0.007) eta 0:26:33 loss 1.1059 (1.0727) acc 96.8750 (99.6875) lr 0.002600 +FPS@all 801.881, TIME@all 0.319 +epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.0820 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:33 loss 1.1115 (1.0700) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 801.710, TIME@all 0.319 +epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.012) eta 0:26:34 loss 1.1233 (1.0609) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.318 (0.321) data 0.000 (0.006) eta 0:26:34 loss 1.0634 (1.0735) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 801.401, TIME@all 0.319 +epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.1088 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:33 loss 1.0907 (1.0720) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 801.657, TIME@all 0.319 +epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1037 (1.0545) acc 96.8750 (99.8438) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.0605 (1.0687) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 801.444, TIME@all 0.319 +epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1820 (1.0577) acc 96.8750 (99.8438) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.1256 (1.0694) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 801.514, TIME@all 0.319 +epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.0911 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.0937 (1.0702) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 801.565, TIME@all 0.319 +epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1190 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [251/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:26:34 loss 1.0532 (1.0708) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 801.428, TIME@all 0.319 +epoch: [252/350][20/50] time 0.320 (0.313) data 0.001 (0.013) eta 0:25:44 loss 1.1117 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0465 (1.0656) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.355, TIME@all 0.313 +epoch: [252/350][20/50] time 0.322 (0.313) data 0.000 (0.014) eta 0:25:45 loss 1.0802 (1.0615) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0549 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.070, TIME@all 0.313 +epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:45 loss 1.1015 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:25:41 loss 1.0574 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 816.891, TIME@all 0.313 +epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.012) eta 0:25:45 loss 1.0834 (1.0621) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:25:41 loss 1.0647 (1.0654) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.911, TIME@all 0.313 +epoch: [252/350][20/50] time 0.320 (0.313) data 0.001 (0.013) eta 0:25:44 loss 1.0839 (1.0641) acc 100.0000 (99.8438) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0564 (1.0657) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.120, TIME@all 0.313 +epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:44 loss 1.0598 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:25:40 loss 1.0464 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.295, TIME@all 0.313 +epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:45 loss 1.1003 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:41 loss 1.0509 (1.0652) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.953, TIME@all 0.313 +epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.012) eta 0:25:45 loss 1.0540 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [252/350][40/50] time 0.314 (0.314) data 0.001 (0.006) eta 0:25:41 loss 1.0502 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.028, TIME@all 0.313 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.1424 (1.0624) acc 96.8750 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:25:18 loss 1.0609 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.411, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.0884 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:25:18 loss 1.0782 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.462, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1351 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0509 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.319, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1193 (1.0604) acc 100.0000 (99.8438) lr 0.002600 +epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:25:18 loss 1.0461 (1.0651) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.326, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1024 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:18 loss 1.0870 (1.0635) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.412, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.1591 (1.0633) acc 96.8750 (99.6875) lr 0.002600 +epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0628 (1.0687) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.301, TIME@all 0.312 +epoch: [253/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.0671 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:25:18 loss 1.0535 (1.0668) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.389, TIME@all 0.312 +epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.0889 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0504 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.332, TIME@all 0.312 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.1083 (1.0600) acc 100.0000 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1139 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.651, TIME@all 0.313 +epoch: [254/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:25:16 loss 1.0733 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1092 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 816.554, TIME@all 0.314 +epoch: [254/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.0589 (1.0585) acc 100.0000 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1330 (1.0720) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 816.490, TIME@all 0.314 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0827 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1054 (1.0697) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 816.488, TIME@all 0.314 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.0684 (1.0654) acc 100.0000 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1138 (1.0722) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.557, TIME@all 0.314 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.1135 (1.0624) acc 96.8750 (99.8438) lr 0.002600 +epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1502 (1.0741) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 816.516, TIME@all 0.314 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0672 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.0636 (1.0730) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.535, TIME@all 0.314 +epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0532 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.0540 (1.0669) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 816.540, TIME@all 0.314 +epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:24:57 loss 1.0626 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0632 (1.0659) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.155, TIME@all 0.312 +epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0523 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0553 (1.0667) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.082, TIME@all 0.312 +epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:24:58 loss 1.0600 (1.0601) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:24:48 loss 1.0695 (1.0635) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.031, TIME@all 0.312 +epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.1020 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0670 (1.0668) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.021, TIME@all 0.312 +epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0673 (1.0612) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0956 (1.0650) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.059, TIME@all 0.312 +epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:57 loss 1.0645 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0833 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.111, TIME@all 0.312 +epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:57 loss 1.1141 (1.0665) acc 100.0000 (99.8438) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.1512 (1.0690) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 820.161, TIME@all 0.312 +epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0571 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.1289 (1.0627) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 820.030, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0828 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0571 (1.0674) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.417, TIME@all 0.312 +epoch: [256/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:24:35 loss 1.0620 (1.0514) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:24:31 loss 1.0583 (1.0679) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.478, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:24:36 loss 1.0954 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0637 (1.0670) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.341, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0821 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0586 (1.0629) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.303, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0675 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0582 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.352, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0529 (1.0532) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0596 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.368, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0760 (1.0599) acc 100.0000 (99.8438) lr 0.002600 +epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0478 (1.0678) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.409, TIME@all 0.312 +epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0821 (1.0528) acc 100.0000 (100.0000) lr 0.002600 +epoch: [256/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 0:24:31 loss 1.0600 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.365, TIME@all 0.312 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:24:21 loss 1.0877 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0563 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 815.516, TIME@all 0.314 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.0779 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0624 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.394, TIME@all 0.314 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:24:21 loss 1.0566 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.006) eta 0:24:23 loss 1.0780 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.413, TIME@all 0.314 +epoch: [257/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:24:21 loss 1.1451 (1.0599) acc 100.0000 (99.8438) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.006) eta 0:24:23 loss 1.0867 (1.0677) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 815.428, TIME@all 0.314 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.1386 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0712 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.395, TIME@all 0.314 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.1191 (1.0611) acc 96.8750 (99.6875) lr 0.002600 +epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.1240 (1.0669) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 815.458, TIME@all 0.314 +epoch: [257/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 0:24:21 loss 1.0545 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0727 (1.0670) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.429, TIME@all 0.314 +epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.0909 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +epoch: [257/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0479 (1.0671) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 815.481, TIME@all 0.314 +epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0594 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0747 (1.0665) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.566, TIME@all 0.312 +epoch: [258/350][20/50] time 0.310 (0.315) data 0.000 (0.013) eta 0:24:16 loss 1.1340 (1.0627) acc 96.8750 (99.5312) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:24:04 loss 1.1445 (1.0709) acc 96.8750 (99.6094) lr 0.002600 +FPS@all 819.596, TIME@all 0.312 +epoch: [258/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0743 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0574 (1.0656) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.485, TIME@all 0.312 +epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.1720 (1.0620) acc 96.8750 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0951 (1.0733) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.470, TIME@all 0.312 +epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0771 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0711 (1.0714) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.536, TIME@all 0.312 +epoch: [258/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.1036 (1.0641) acc 100.0000 (99.8438) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0844 (1.0700) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.531, TIME@all 0.312 +epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0610 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0750 (1.0709) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.518, TIME@all 0.312 +epoch: [258/350][20/50] time 0.311 (0.315) data 0.001 (0.012) eta 0:24:16 loss 1.1189 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [258/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:24:04 loss 1.0748 (1.0695) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.600, TIME@all 0.312 +epoch: [259/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1706 (1.0609) acc 96.8750 (99.8438) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0572 (1.0665) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.678, TIME@all 0.312 +epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1602 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0582 (1.0730) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.593, TIME@all 0.312 +epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:23:49 loss 1.0685 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:23:45 loss 1.0508 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.538, TIME@all 0.312 +epoch: [259/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1129 (1.0563) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0507 (1.0683) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.554, TIME@all 0.312 +epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1173 (1.0568) acc 96.8750 (99.8438) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0715 (1.0684) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.586, TIME@all 0.312 +epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1403 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.001 (0.007) eta 0:23:45 loss 1.0662 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.527, TIME@all 0.312 +epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.0716 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0469 (1.0662) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.554, TIME@all 0.312 +epoch: [259/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:23:49 loss 1.1356 (1.0608) acc 96.8750 (99.8438) lr 0.002600 +epoch: [259/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0481 (1.0704) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.603, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0691 (1.0622) acc 100.0000 (99.8438) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.1186 (1.0757) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.604, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0615 (1.0581) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0517 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.472, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0641 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0614 (1.0674) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.458, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:23:37 loss 1.0904 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0483 (1.0669) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.487, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:23:37 loss 1.1344 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0495 (1.0750) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.470, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.1105 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0516 (1.0667) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.464, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0795 (1.0646) acc 100.0000 (99.8438) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0617 (1.0749) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 819.558, TIME@all 0.312 +epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.1173 (1.0619) acc 100.0000 (99.8438) lr 0.002600 +epoch: [260/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:23:31 loss 1.0524 (1.0725) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.514, TIME@all 0.312 +epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:23:18 loss 1.0692 (1.0615) acc 100.0000 (99.8438) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:23:13 loss 1.0502 (1.0681) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.083, TIME@all 0.312 +epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0695 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0659 (1.0670) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.127, TIME@all 0.312 +epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0671 (1.0672) acc 100.0000 (99.6875) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0801 (1.0720) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.035, TIME@all 0.312 +epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0844 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0762 (1.0691) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.984, TIME@all 0.312 +epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:18 loss 1.0695 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:23:13 loss 1.0735 (1.0630) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.033, TIME@all 0.312 +epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0536 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0920 (1.0732) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.046, TIME@all 0.312 +epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0677 (1.0600) acc 100.0000 (99.8438) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0602 (1.0699) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.071, TIME@all 0.312 +epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0597 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0695 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.035, TIME@all 0.312 +epoch: [262/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:23:13 loss 1.0680 (1.0675) acc 100.0000 (99.6875) lr 0.002600 +epoch: [262/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0673 (1.0730) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 816.949, TIME@all 0.313 +epoch: [262/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:23:13 loss 1.0664 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:23:04 loss 1.0717 (1.0629) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.966, TIME@all 0.313 +epoch: [262/350][20/50] time 0.315 (0.314) data 0.001 (0.011) eta 0:23:13 loss 1.0466 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0574 (1.0643) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 816.874, TIME@all 0.313 +epoch: [262/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:23:13 loss 1.0533 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0794 (1.0703) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.830, TIME@all 0.313 +epoch: [262/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:23:13 loss 1.0502 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:23:04 loss 1.1793 (1.0656) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.909, TIME@all 0.313 +epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0553 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0652 (1.0648) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 816.891, TIME@all 0.313 +epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0638 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.001 (0.006) eta 0:23:04 loss 1.0613 (1.0606) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 816.928, TIME@all 0.313 +epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0905 (1.0614) acc 100.0000 (99.8438) lr 0.002600 +epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0801 (1.0688) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 816.984, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.014) eta 0:22:53 loss 1.0478 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.1144 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.417, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.012) eta 0:22:54 loss 1.0482 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:22:47 loss 1.0690 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.223, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:22:54 loss 1.0548 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0658 (1.0661) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.215, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:22:53 loss 1.0811 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0845 (1.0675) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.283, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0760 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.1074 (1.0665) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 818.319, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.014) eta 0:22:53 loss 1.0564 (1.0600) acc 100.0000 (99.8438) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0836 (1.0713) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.278, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0689 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0669 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.288, TIME@all 0.313 +epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0503 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0861 (1.0685) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.336, TIME@all 0.313 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0551 (1.0545) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0475 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.299, TIME@all 0.312 +epoch: [264/350][20/50] time 0.314 (0.313) data 0.000 (0.011) eta 0:22:35 loss 1.0502 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0619 (1.0648) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.197, TIME@all 0.313 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0464 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0677 (1.0705) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.158, TIME@all 0.313 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0458 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0559 (1.0680) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.223, TIME@all 0.312 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0570 (1.0594) acc 100.0000 (99.8438) lr 0.002600 +epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0549 (1.0710) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.243, TIME@all 0.312 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0564 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:22:29 loss 1.0995 (1.0695) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.242, TIME@all 0.312 +epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0785 (1.0581) acc 100.0000 (99.8438) lr 0.002600 +epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0504 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.199, TIME@all 0.313 +epoch: [264/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0548 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0454 (1.0715) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.224, TIME@all 0.312 +epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0618 (1.0570) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0516 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.503, TIME@all 0.313 +epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0676 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0492 (1.0628) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.426, TIME@all 0.313 +epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0585 (1.0628) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0523 (1.0653) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.356, TIME@all 0.313 +epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0600 (1.0644) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0542 (1.0650) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.378, TIME@all 0.313 +epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0505 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0529 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 817.313, TIME@all 0.313 +epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0605 (1.0580) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0548 (1.0674) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.421, TIME@all 0.313 +epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0611 (1.0662) acc 100.0000 (99.8438) lr 0.002600 +epoch: [265/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0521 (1.0734) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 817.364, TIME@all 0.313 +epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0555 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [265/350][40/50] time 0.315 (0.314) data 0.001 (0.006) eta 0:22:16 loss 1.0515 (1.0684) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 817.406, TIME@all 0.313 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:22:12 loss 1.1244 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.1069 (1.0718) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.605, TIME@all 0.314 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0599 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.1140 (1.0666) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 814.574, TIME@all 0.314 +epoch: [266/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:22:11 loss 1.0751 (1.0570) acc 100.0000 (99.8438) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.0579 (1.0650) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.663, TIME@all 0.314 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:22:12 loss 1.1002 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +epoch: [266/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:22:05 loss 1.0522 (1.0684) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.505, TIME@all 0.314 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0965 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.1032 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 814.578, TIME@all 0.314 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 0:22:12 loss 1.1078 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.0783 (1.0676) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 814.596, TIME@all 0.314 +epoch: [266/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0937 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.001 (0.006) eta 0:22:04 loss 1.0824 (1.0755) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 814.607, TIME@all 0.314 +epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0874 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.0765 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 814.570, TIME@all 0.314 +epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0695 (1.0659) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:21:42 loss 1.0613 (1.0653) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.436, TIME@all 0.313 +epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0742 (1.0608) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:42 loss 1.0592 (1.0627) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.466, TIME@all 0.313 +epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.1049 (1.0601) acc 96.8750 (99.6875) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0636 (1.0676) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 818.368, TIME@all 0.313 +epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0772 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0653 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.294, TIME@all 0.313 +epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0518 (1.0575) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0827 (1.0663) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.353, TIME@all 0.313 +epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0669 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:21:43 loss 1.0578 (1.0668) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.366, TIME@all 0.313 +epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0691 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0736 (1.0654) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.302, TIME@all 0.313 +epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0807 (1.0644) acc 100.0000 (99.8438) lr 0.002600 +epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.1050 (1.0686) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.350, TIME@all 0.313 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0613 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0810 (1.0659) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.840, TIME@all 0.312 +epoch: [268/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:21:30 loss 1.0567 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:25 loss 1.0880 (1.0674) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.902, TIME@all 0.312 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0761 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0644 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.785, TIME@all 0.312 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0495 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.1158 (1.0645) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.783, TIME@all 0.312 +epoch: [268/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:21:31 loss 1.0670 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:21:25 loss 1.0569 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.760, TIME@all 0.312 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:21:30 loss 1.1609 (1.0594) acc 96.8750 (99.8438) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:25 loss 1.0619 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.762, TIME@all 0.312 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0718 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0844 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.815, TIME@all 0.312 +epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0970 (1.0642) acc 100.0000 (100.0000) lr 0.002600 +epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0788 (1.0717) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.825, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0573 (1.0604) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.1133 (1.0700) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.623, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:21:19 loss 1.0424 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0952 (1.0706) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 819.686, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:21:19 loss 1.0801 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:21:11 loss 1.0995 (1.0741) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.569, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0838 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0917 (1.0722) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.546, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0562 (1.0549) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.1012 (1.0730) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.568, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:21:19 loss 1.1331 (1.0623) acc 100.0000 (100.0000) lr 0.002600 +epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0828 (1.0733) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.564, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0806 (1.0638) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0709 (1.0724) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.620, TIME@all 0.312 +epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0717 (1.0585) acc 100.0000 (99.8438) lr 0.002600 +epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0773 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.617, TIME@all 0.312 +epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:21:04 loss 1.0617 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0585 (1.0638) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.399, TIME@all 0.313 +epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0463 (1.0606) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0931 (1.0691) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.132, TIME@all 0.313 +epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0555 (1.0577) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0492 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.170, TIME@all 0.313 +epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0477 (1.0542) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0641 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.142, TIME@all 0.313 +epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0490 (1.0581) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0955 (1.0644) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 818.136, TIME@all 0.313 +epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.011) eta 0:21:05 loss 1.0888 (1.0628) acc 100.0000 (99.8438) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0695 (1.0637) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.085, TIME@all 0.313 +epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0670 (1.0517) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0807 (1.0652) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.141, TIME@all 0.313 +epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0527 (1.0582) acc 100.0000 (100.0000) lr 0.002600 +epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0576 (1.0674) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.175, TIME@all 0.313 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:20:46 loss 1.1241 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:39 loss 1.0702 (1.0750) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.792, TIME@all 0.312 +epoch: [271/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1162 (1.0653) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0894 (1.0752) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.639, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0583 (1.0664) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.1296 (1.0753) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.756, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1315 (1.0689) acc 100.0000 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.1048 (1.0793) acc 100.0000 (99.6094) lr 0.002600 +FPS@all 819.742, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1090 (1.0607) acc 96.8750 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0753 (1.0682) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.635, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:20:46 loss 1.1258 (1.0642) acc 96.8750 (99.8438) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:20:39 loss 1.1463 (1.0707) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 819.666, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0915 (1.0655) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0688 (1.0763) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.657, TIME@all 0.312 +epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0746 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +epoch: [271/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:20:39 loss 1.0872 (1.0717) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.686, TIME@all 0.312 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:20:30 loss 1.0575 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:24 loss 1.1060 (1.0669) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.684, TIME@all 0.313 +epoch: [272/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:20:30 loss 1.0574 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0777 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.592, TIME@all 0.313 +epoch: [272/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:20:30 loss 1.0462 (1.0523) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:20:25 loss 1.0629 (1.0640) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.553, TIME@all 0.313 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:31 loss 1.0642 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0622 (1.0693) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.576, TIME@all 0.313 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:20:31 loss 1.0633 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:20:25 loss 1.0603 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.549, TIME@all 0.313 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:20:30 loss 1.0610 (1.0555) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0566 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.593, TIME@all 0.313 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:31 loss 1.0653 (1.0506) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0546 (1.0643) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.551, TIME@all 0.313 +epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:30 loss 1.0518 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0766 (1.0717) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.618, TIME@all 0.313 +epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0759 (1.0619) acc 100.0000 (99.8438) lr 0.002600 +epoch: [273/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0737 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.507, TIME@all 0.312 +epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.1392 (1.0651) acc 100.0000 (99.8438) lr 0.002600 +epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0632 (1.0664) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.347, TIME@all 0.312 +epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:20:16 loss 1.0945 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:20:07 loss 1.0621 (1.0627) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.330, TIME@all 0.312 +epoch: [273/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:20:16 loss 1.0897 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0706 (1.0731) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.523, TIME@all 0.312 +epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0774 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0532 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.392, TIME@all 0.312 +epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0528 (1.0566) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:20:07 loss 1.0507 (1.0687) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.373, TIME@all 0.312 +epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0982 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0783 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.355, TIME@all 0.312 +epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0677 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0610 (1.0630) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.406, TIME@all 0.312 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0670 (1.0604) acc 100.0000 (99.8438) lr 0.002600 +epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0788 (1.0725) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.213, TIME@all 0.312 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:20:04 loss 1.0545 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:19:54 loss 1.0659 (1.0619) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.286, TIME@all 0.312 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0503 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0633 (1.0660) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.232, TIME@all 0.312 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0587 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0555 (1.0608) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.146, TIME@all 0.313 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0682 (1.0626) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0534 (1.0709) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.198, TIME@all 0.313 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0677 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0758 (1.0637) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.193, TIME@all 0.313 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.1912 (1.0649) acc 96.8750 (99.8438) lr 0.002600 +epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0607 (1.0702) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.131, TIME@all 0.313 +epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0484 (1.0572) acc 100.0000 (100.0000) lr 0.002600 +epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0498 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.197, TIME@all 0.313 +epoch: [275/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:19:41 loss 1.0695 (1.0611) acc 100.0000 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.0522 (1.0692) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.821, TIME@all 0.312 +epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.1124 (1.0592) acc 100.0000 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.1097 (1.0676) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.696, TIME@all 0.312 +epoch: [275/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.0742 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.1008 (1.0768) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.732, TIME@all 0.312 +epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:19:41 loss 1.1003 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.0675 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.694, TIME@all 0.312 +epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.1078 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.0603 (1.0720) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.707, TIME@all 0.312 +epoch: [275/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:19:41 loss 1.0933 (1.0527) acc 100.0000 (100.0000) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.1168 (1.0638) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.690, TIME@all 0.312 +epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.0581 (1.0622) acc 100.0000 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:19:36 loss 1.1114 (1.0721) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.696, TIME@all 0.312 +epoch: [275/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:19:41 loss 1.1303 (1.0556) acc 96.8750 (99.8438) lr 0.002600 +epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.0798 (1.0676) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.748, TIME@all 0.312 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0923 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:19:22 loss 1.2261 (1.0761) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.666, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:19:28 loss 1.0837 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0640 (1.0703) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.536, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0616 (1.0611) acc 100.0000 (99.8438) lr 0.002600 +epoch: [276/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:19:22 loss 1.0784 (1.0693) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.541, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0591 (1.0577) acc 100.0000 (99.8438) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0673 (1.0730) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.596, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:19:28 loss 1.0544 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0556 (1.0718) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.563, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0478 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0780 (1.0716) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.565, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0744 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [276/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:22 loss 1.0914 (1.0664) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.551, TIME@all 0.313 +epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0713 (1.0644) acc 100.0000 (99.8438) lr 0.002600 +epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0776 (1.0699) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.562, TIME@all 0.313 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0594 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0979 (1.0666) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 822.018, TIME@all 0.311 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0481 (1.0608) acc 100.0000 (99.8438) lr 0.002600 +epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.1277 (1.0738) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 821.925, TIME@all 0.311 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0613 (1.0556) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.0785 (1.0699) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 821.883, TIME@all 0.311 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.001 (0.012) eta 0:19:09 loss 1.0631 (1.0647) acc 100.0000 (99.8438) lr 0.002600 +epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.0530 (1.0717) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.909, TIME@all 0.311 +epoch: [277/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0575 (1.0591) acc 100.0000 (99.8438) lr 0.002600 +epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0754 (1.0680) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 821.949, TIME@all 0.311 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0705 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0967 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.891, TIME@all 0.311 +epoch: [277/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0577 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.1134 (1.0723) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.928, TIME@all 0.311 +epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0645 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0854 (1.0712) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.893, TIME@all 0.311 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.1210 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0530 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.969, TIME@all 0.312 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:18:59 loss 1.0819 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0509 (1.0647) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.998, TIME@all 0.312 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.0808 (1.0551) acc 100.0000 (99.8438) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0925 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.858, TIME@all 0.312 +epoch: [278/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:18:59 loss 1.1253 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:18:50 loss 1.0595 (1.0692) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.860, TIME@all 0.312 +epoch: [278/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:18:58 loss 1.0933 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:18:50 loss 1.0620 (1.0676) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.029, TIME@all 0.312 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.0953 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0628 (1.0668) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.915, TIME@all 0.312 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.1094 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0458 (1.0640) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.889, TIME@all 0.312 +epoch: [278/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:18:59 loss 1.0767 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0618 (1.0681) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.946, TIME@all 0.312 +epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0765 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1777 (1.0664) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.105, TIME@all 0.312 +epoch: [279/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0535 (1.0525) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1228 (1.0627) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.024, TIME@all 0.312 +epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:18:38 loss 1.0546 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0893 (1.0621) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.021, TIME@all 0.312 +epoch: [279/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:38 loss 1.0774 (1.0575) acc 100.0000 (99.8438) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0872 (1.0624) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.052, TIME@all 0.312 +epoch: [279/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:18:38 loss 1.0975 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:18:31 loss 1.0819 (1.0667) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.184, TIME@all 0.312 +epoch: [279/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:18:38 loss 1.0859 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0964 (1.0647) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.061, TIME@all 0.312 +epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0646 (1.0531) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1593 (1.0637) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 821.109, TIME@all 0.312 +epoch: [279/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0609 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0751 (1.0607) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.087, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0521 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0577 (1.0743) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.749, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:18:21 loss 1.0596 (1.0615) acc 100.0000 (99.8438) lr 0.002600 +epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0569 (1.0703) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.815, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0778 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0565 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.681, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0571 (1.0535) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0599 (1.0657) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.711, TIME@all 0.312 +epoch: [280/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:21 loss 1.0560 (1.0581) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:15 loss 1.0485 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.682, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0580 (1.0522) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0540 (1.0675) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.730, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:18:21 loss 1.0631 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0545 (1.0622) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.767, TIME@all 0.312 +epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.1517 (1.0622) acc 96.8750 (99.6875) lr 0.002600 +epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0537 (1.0698) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.713, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0469 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0628 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.239, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:07 loss 1.0633 (1.0662) acc 100.0000 (100.0000) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0457 (1.0681) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.311, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0559 (1.0644) acc 100.0000 (99.6875) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0610 (1.0667) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.142, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0591 (1.0623) acc 100.0000 (99.6875) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0541 (1.0717) acc 100.0000 (99.3750) lr 0.002600 +FPS@all 820.157, TIME@all 0.312 +epoch: [281/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:07 loss 1.1096 (1.0620) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:18:02 loss 1.0477 (1.0698) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.266, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0951 (1.0616) acc 96.8750 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0447 (1.0700) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.169, TIME@all 0.312 +epoch: [281/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0570 (1.0601) acc 100.0000 (99.8438) lr 0.002600 +epoch: [281/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:18:02 loss 1.0536 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.212, TIME@all 0.312 +epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0600 (1.0605) acc 100.0000 (100.0000) lr 0.002600 +epoch: [281/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0842 (1.0700) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.188, TIME@all 0.312 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1194 (1.0590) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.001 (0.007) eta 0:17:42 loss 1.0724 (1.0677) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.941, TIME@all 0.311 +epoch: [282/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.0936 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:41 loss 1.0569 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.983, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:49 loss 1.1055 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0653 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.825, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:49 loss 1.0928 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0479 (1.0635) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 822.828, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.0565 (1.0557) acc 100.0000 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0462 (1.0635) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.889, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:48 loss 1.0620 (1.0608) acc 100.0000 (99.8438) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0525 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.902, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1407 (1.0634) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0713 (1.0668) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 822.890, TIME@all 0.311 +epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1288 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0735 (1.0643) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 822.897, TIME@all 0.311 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0518 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0651 (1.0669) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.229, TIME@all 0.312 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.1001 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0567 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.254, TIME@all 0.312 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0466 (1.0562) acc 100.0000 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0565 (1.0704) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.075, TIME@all 0.312 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0579 (1.0552) acc 100.0000 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0935 (1.0658) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.112, TIME@all 0.312 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.011) eta 0:17:36 loss 1.0707 (1.0560) acc 100.0000 (99.8438) lr 0.002600 +epoch: [283/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.1615 (1.0706) acc 96.8750 (99.6094) lr 0.002600 +FPS@all 820.099, TIME@all 0.312 +epoch: [283/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0523 (1.0547) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0554 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.173, TIME@all 0.312 +epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0861 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:17:30 loss 1.0867 (1.0699) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.208, TIME@all 0.312 +epoch: [283/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0612 (1.0530) acc 100.0000 (100.0000) lr 0.002600 +epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0764 (1.0644) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.135, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1092 (1.0571) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:14 loss 1.1271 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.465, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:17:21 loss 1.0694 (1.0610) acc 100.0000 (99.8438) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:17:14 loss 1.1863 (1.0710) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 819.471, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.2068 (1.0728) acc 96.8750 (99.6875) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.2257 (1.0795) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 819.347, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1417 (1.0599) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1486 (1.0671) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.367, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.0849 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1821 (1.0745) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 819.408, TIME@all 0.312 +epoch: [284/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1134 (1.0574) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:17:14 loss 1.1131 (1.0753) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.430, TIME@all 0.312 +epoch: [284/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.0892 (1.0607) acc 100.0000 (99.6875) lr 0.002600 +epoch: [284/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1210 (1.0711) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.349, TIME@all 0.312 +epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1210 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1240 (1.0779) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 819.375, TIME@all 0.312 +epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.1203 (1.0641) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:16:57 loss 1.0450 (1.0704) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.557, TIME@all 0.312 +epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:17:05 loss 1.1627 (1.0613) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:58 loss 1.0519 (1.0693) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.446, TIME@all 0.312 +epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0890 (1.0629) acc 100.0000 (99.6875) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:16:58 loss 1.0560 (1.0732) acc 100.0000 (99.5312) lr 0.002600 +FPS@all 820.424, TIME@all 0.312 +epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0964 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:16:57 loss 1.0795 (1.0669) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.498, TIME@all 0.312 +epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0668 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [285/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:16:57 loss 1.0513 (1.0666) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.491, TIME@all 0.312 +epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.1674 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:58 loss 1.0611 (1.0703) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.462, TIME@all 0.312 +epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0627 (1.0648) acc 100.0000 (99.8438) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:16:58 loss 1.0511 (1.0669) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.448, TIME@all 0.312 +epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:17:05 loss 1.2219 (1.0689) acc 93.7500 (99.6875) lr 0.002600 +epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:57 loss 1.0556 (1.0706) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.484, TIME@all 0.312 +epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:16:48 loss 1.0589 (1.0507) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0555 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.986, TIME@all 0.312 +epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1288 (1.0586) acc 100.0000 (99.6875) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0534 (1.0629) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.822, TIME@all 0.312 +epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:16:48 loss 1.0799 (1.0533) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0492 (1.0644) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.820, TIME@all 0.312 +epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.0965 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0689 (1.0635) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.885, TIME@all 0.312 +epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1324 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0533 (1.0667) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.905, TIME@all 0.312 +epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.0627 (1.0525) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0477 (1.0611) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.870, TIME@all 0.312 +epoch: [286/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1191 (1.0616) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0537 (1.0652) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.880, TIME@all 0.312 +epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:48 loss 1.0989 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0616 (1.0649) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.888, TIME@all 0.312 +epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0584 (1.0579) acc 100.0000 (99.8438) lr 0.002600 +epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0540 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 809.534, TIME@all 0.316 +epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.1104 (1.0629) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0869 (1.0720) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 809.587, TIME@all 0.316 +epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0730 (1.0558) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0636 (1.0688) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 809.411, TIME@all 0.316 +epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0537 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0583 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 809.433, TIME@all 0.316 +epoch: [287/350][20/50] time 0.326 (0.317) data 0.001 (0.012) eta 0:16:48 loss 1.0844 (1.0545) acc 96.8750 (99.8438) lr 0.002600 +epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0727 (1.0621) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 809.681, TIME@all 0.316 +epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.013) eta 0:16:48 loss 1.0773 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.007) eta 0:16:44 loss 1.1349 (1.0715) acc 96.8750 (99.6875) lr 0.002600 +FPS@all 809.476, TIME@all 0.316 +epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0592 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.309 (0.318) data 0.001 (0.006) eta 0:16:45 loss 1.0626 (1.0639) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 809.463, TIME@all 0.316 +epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0545 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0549 (1.0602) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 809.463, TIME@all 0.316 +epoch: [288/350][20/50] time 0.315 (0.312) data 0.001 (0.013) eta 0:16:18 loss 1.0604 (1.0586) acc 100.0000 (99.8438) lr 0.002600 +epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0494 (1.0633) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.770, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1134 (1.0587) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0448 (1.0679) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.545, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.001 (0.012) eta 0:16:18 loss 1.0775 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0492 (1.0636) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.584, TIME@all 0.312 +epoch: [288/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:16:18 loss 1.0500 (1.0567) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:16:13 loss 1.0692 (1.0651) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.379, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1051 (1.0560) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0424 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.397, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:16:18 loss 1.0759 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:16:13 loss 1.0611 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.419, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1076 (1.0592) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0463 (1.0645) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.404, TIME@all 0.312 +epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.0751 (1.0632) acc 100.0000 (100.0000) lr 0.002600 +epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0564 (1.0675) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.426, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.1081 (1.0575) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.0853 (1.0677) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.181, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.0888 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:15:56 loss 1.1359 (1.0701) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.214, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:02 loss 1.0776 (1.0581) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.1046 (1.0671) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.054, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.1520 (1.0639) acc 100.0000 (99.8438) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.1033 (1.0726) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.091, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.0844 (1.0561) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:15:56 loss 1.1669 (1.0658) acc 96.8750 (99.8438) lr 0.002600 +FPS@all 820.205, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.0908 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:15:56 loss 1.0621 (1.0648) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.166, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.0913 (1.0569) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:15:56 loss 1.0615 (1.0660) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.105, TIME@all 0.312 +epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.1225 (1.0578) acc 100.0000 (100.0000) lr 0.002600 +epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:15:56 loss 1.0573 (1.0663) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.135, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0626 (1.0596) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0522 (1.0683) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.355, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0550 (1.0538) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0519 (1.0618) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 821.415, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.011) eta 0:15:46 loss 1.0574 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0579 (1.0686) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 821.261, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0723 (1.0554) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0806 (1.0708) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.242, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:45 loss 1.0434 (1.0553) acc 100.0000 (99.8438) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0502 (1.0694) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.399, TIME@all 0.312 +epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0447 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0469 (1.0670) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.293, TIME@all 0.312 +epoch: [290/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:15:46 loss 1.0732 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:15:39 loss 1.0470 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.264, TIME@all 0.312 +epoch: [290/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0734 (1.0588) acc 100.0000 (100.0000) lr 0.002600 +epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0717 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 821.367, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0774 (1.0686) acc 100.0000 (99.6875) lr 0.002600 +epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.0667 (1.0695) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.563, TIME@all 0.312 +epoch: [291/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0631 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1014 (1.0654) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.489, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1826 (1.0606) acc 96.8750 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1119 (1.0726) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.444, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:15:31 loss 1.0806 (1.0544) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:15:25 loss 1.1424 (1.0671) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.457, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1142 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1173 (1.0681) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 820.501, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1349 (1.0587) acc 96.8750 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:15:25 loss 1.0753 (1.0643) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.491, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0536 (1.0608) acc 100.0000 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:15:25 loss 1.1294 (1.0729) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.457, TIME@all 0.312 +epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0645 (1.0566) acc 100.0000 (99.8438) lr 0.002600 +epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.0740 (1.0668) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.480, TIME@all 0.312 +epoch: [292/350][20/50] time 0.318 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0825 (1.0589) acc 100.0000 (99.8438) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:15:12 loss 1.0515 (1.0698) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.191, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0834 (1.0598) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.1003 (1.0759) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.246, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.0626 (1.0502) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0442 (1.0636) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.112, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.001 (0.012) eta 0:15:15 loss 1.0747 (1.0539) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0721 (1.0679) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.122, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.001 (0.013) eta 0:15:15 loss 1.0679 (1.0541) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:15:12 loss 1.0752 (1.0695) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.133, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0880 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:15:12 loss 1.0434 (1.0685) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.137, TIME@all 0.313 +epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.0664 (1.0597) acc 100.0000 (100.0000) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0588 (1.0733) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.252, TIME@all 0.313 +epoch: [292/350][20/50] time 0.318 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.1226 (1.0619) acc 100.0000 (99.8438) lr 0.002600 +epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:15:12 loss 1.0612 (1.0683) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.183, TIME@all 0.313 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:14:58 loss 1.0579 (1.0624) acc 100.0000 (99.8438) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.1036 (1.0694) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.797, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0569 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0873 (1.0709) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.707, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:14:59 loss 1.0576 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:14:53 loss 1.0811 (1.0706) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.633, TIME@all 0.312 +epoch: [293/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0972 (1.0589) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0651 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.586, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0472 (1.0552) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0590 (1.0662) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.654, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0889 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0953 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.735, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0517 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0549 (1.0701) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.676, TIME@all 0.312 +epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0607 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0711 (1.0685) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.672, TIME@all 0.312 +epoch: [294/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0592 (1.0559) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:38 loss 1.0685 (1.0646) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.661, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0520 (1.0515) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:38 loss 1.0565 (1.0579) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.673, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:14:47 loss 1.0537 (1.0576) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:14:39 loss 1.0503 (1.0661) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.583, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:14:47 loss 1.0724 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:14:39 loss 1.0793 (1.0610) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.586, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0788 (1.0562) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0484 (1.0635) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.540, TIME@all 0.313 +epoch: [294/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0433 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0612 (1.0677) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.632, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0665 (1.0534) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0652 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.584, TIME@all 0.313 +epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0542 (1.0586) acc 100.0000 (100.0000) lr 0.002600 +epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0868 (1.0674) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.590, TIME@all 0.313 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0479 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0747 (1.0691) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.536, TIME@all 0.312 +epoch: [295/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0520 (1.0600) acc 100.0000 (99.6875) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0884 (1.0719) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 820.393, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0569 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.1006 (1.0652) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.487, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0480 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.1213 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 820.456, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:14:30 loss 1.0518 (1.0571) acc 100.0000 (99.8438) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:14:23 loss 1.0800 (1.0664) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 820.444, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0525 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0976 (1.0705) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.493, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0457 (1.0553) acc 100.0000 (100.0000) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0644 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 820.544, TIME@all 0.312 +epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0457 (1.0598) acc 100.0000 (99.6875) lr 0.002600 +epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0938 (1.0703) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 820.446, TIME@all 0.312 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1829 (1.0609) acc 100.0000 (99.8438) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.001 (0.006) eta 0:14:03 loss 1.1275 (1.0726) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 824.548, TIME@all 0.310 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:14:10 loss 1.1264 (1.0622) acc 100.0000 (100.0000) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.313 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0700 (1.0699) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 824.579, TIME@all 0.310 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1059 (1.0600) acc 100.0000 (100.0000) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0415 (1.0675) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 824.471, TIME@all 0.311 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:11 loss 1.1242 (1.0593) acc 100.0000 (100.0000) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0494 (1.0634) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 824.453, TIME@all 0.311 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:11 loss 1.1269 (1.0628) acc 100.0000 (99.6875) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0658 (1.0661) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 824.498, TIME@all 0.310 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:14:11 loss 1.1366 (1.0615) acc 96.8750 (99.8438) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0574 (1.0706) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 824.481, TIME@all 0.310 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:14:11 loss 1.1480 (1.0591) acc 100.0000 (100.0000) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.313 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0515 (1.0649) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 824.512, TIME@all 0.310 +epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1488 (1.0603) acc 100.0000 (100.0000) lr 0.002600 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [296/350][40/50] time 0.312 (0.311) data 0.001 (0.006) eta 0:14:03 loss 1.0670 (1.0728) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 824.541, TIME@all 0.310 +epoch: [297/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0645 (1.0573) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0648 (1.0695) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.727, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0675 (1.0597) acc 100.0000 (99.8438) lr 0.002600 +epoch: [297/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0681 (1.0689) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.617, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0707 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0918 (1.0688) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 818.642, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.011) eta 0:13:59 loss 1.0804 (1.0524) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.1046 (1.0661) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 818.612, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:13:59 loss 1.0835 (1.0537) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:13:52 loss 1.0817 (1.0620) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.876, TIME@all 0.313 +epoch: [297/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0749 (1.0522) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:13:52 loss 1.1059 (1.0618) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 818.647, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0722 (1.0557) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0955 (1.0657) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.661, TIME@all 0.313 +epoch: [297/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 0:13:59 loss 1.0673 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0995 (1.0660) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.700, TIME@all 0.313 +epoch: [298/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:13:41 loss 1.0528 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1882 (1.0699) acc 96.8750 (99.9219) lr 0.002600 +FPS@all 823.308, TIME@all 0.311 +epoch: [298/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.0604 (1.0649) acc 100.0000 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.1367 (1.0729) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 823.126, TIME@all 0.311 +epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:13:41 loss 1.0492 (1.0543) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.1494 (1.0747) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 823.154, TIME@all 0.311 +epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:13:41 loss 1.0649 (1.0604) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.0915 (1.0682) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 823.207, TIME@all 0.311 +epoch: [298/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:13:41 loss 1.0546 (1.0594) acc 100.0000 (100.0000) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1012 (1.0664) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 823.237, TIME@all 0.311 +epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:13:41 loss 1.0549 (1.0659) acc 100.0000 (99.5312) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1090 (1.0710) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 823.160, TIME@all 0.311 +epoch: [298/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.0591 (1.0607) acc 100.0000 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.0902 (1.0655) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 823.251, TIME@all 0.311 +epoch: [298/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.1372 (1.0650) acc 96.8750 (99.8438) lr 0.002600 +epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.0714 (1.0722) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 823.184, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:13:24 loss 1.0928 (1.0589) acc 100.0000 (99.8438) lr 0.002600 +epoch: [299/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1157 (1.0740) acc 96.8750 (99.7656) lr 0.002600 +FPS@all 822.569, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0796 (1.0548) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0625 (1.0675) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 822.372, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0635 (1.0545) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1252 (1.0700) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 822.412, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.1103 (1.0580) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0654 (1.0797) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 822.477, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 0:13:24 loss 1.1204 (1.0568) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.001 (0.006) eta 0:13:17 loss 1.0739 (1.0684) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 822.432, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.1289 (1.0565) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0868 (1.0724) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 822.389, TIME@all 0.311 +epoch: [299/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:13:24 loss 1.1288 (1.0655) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0712 (1.0779) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 822.516, TIME@all 0.311 +epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0843 (1.0540) acc 100.0000 (100.0000) lr 0.002600 +epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1382 (1.0768) acc 100.0000 (99.7656) lr 0.002600 +FPS@all 822.469, TIME@all 0.311 +epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0596 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0628 (1.0666) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 819.108, TIME@all 0.313 +epoch: [300/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:13:13 loss 1.0819 (1.0583) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0673 (1.0696) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.147, TIME@all 0.313 +epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.012) eta 0:13:13 loss 1.0533 (1.0536) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:13:06 loss 1.0753 (1.0651) acc 100.0000 (100.0000) lr 0.002600 +FPS@all 818.984, TIME@all 0.313 +epoch: [300/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:13:13 loss 1.0514 (1.0564) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0500 (1.0737) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.060, TIME@all 0.313 +epoch: [300/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0642 (1.0551) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:07 loss 1.0521 (1.0713) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 818.993, TIME@all 0.313 +epoch: [300/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0551 (1.0526) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:06 loss 1.0448 (1.0654) acc 100.0000 (99.9219) lr 0.002600 +FPS@all 819.045, TIME@all 0.313 +epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0566 (1.0594) acc 100.0000 (99.8438) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.314) data 0.001 (0.007) eta 0:13:06 loss 1.0637 (1.0733) acc 100.0000 (99.6875) lr 0.002600 +FPS@all 819.041, TIME@all 0.313 +epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0602 (1.0585) acc 100.0000 (100.0000) lr 0.002600 +epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:06 loss 1.0499 (1.0710) acc 100.0000 (99.8438) lr 0.002600 +FPS@all 819.070, TIME@all 0.313 +epoch: [301/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0735 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0515 (1.0663) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.410, TIME@all 0.312 +epoch: [301/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0663 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0766 (1.0658) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.209, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:54 loss 1.0454 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:12:48 loss 1.0759 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.278, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0650 (1.0578) acc 100.0000 (99.8438) lr 0.000260 +epoch: [301/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0853 (1.0700) acc 100.0000 (99.4531) lr 0.000260 +FPS@all 821.316, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0501 (1.0526) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0651 (1.0620) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.325, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:12:54 loss 1.0628 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0666 (1.0658) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.276, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0619 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0666 (1.0620) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.234, TIME@all 0.312 +epoch: [301/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:12:54 loss 1.0647 (1.0621) acc 100.0000 (99.6875) lr 0.000260 +epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0601 (1.0660) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.306, TIME@all 0.312 +epoch: [302/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0577 (1.0566) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0846 (1.0652) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.753, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0587 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.1220 (1.0652) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.642, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0575 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0737 (1.0657) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.597, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:12:38 loss 1.0518 (1.0610) acc 100.0000 (99.8438) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:12:33 loss 1.0776 (1.0651) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.688, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0631 (1.0578) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0722 (1.0650) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.719, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.001 (0.012) eta 0:12:38 loss 1.0615 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0926 (1.0676) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.641, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0720 (1.0524) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0742 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.655, TIME@all 0.312 +epoch: [302/350][20/50] time 0.314 (0.312) data 0.001 (0.012) eta 0:12:38 loss 1.0549 (1.0566) acc 100.0000 (100.0000) lr 0.000260 +epoch: [302/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:12:33 loss 1.0646 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.689, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:12:23 loss 1.0797 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0527 (1.0718) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.546, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0752 (1.0639) acc 100.0000 (99.6875) lr 0.000260 +epoch: [303/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:17 loss 1.1225 (1.0759) acc 96.8750 (99.6094) lr 0.000260 +FPS@all 820.381, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.011) eta 0:12:23 loss 1.0771 (1.0582) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0685 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.383, TIME@all 0.312 +epoch: [303/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0716 (1.0641) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.1099 (1.0734) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.453, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0505 (1.0559) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:12:17 loss 1.0893 (1.0697) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.262, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0973 (1.0572) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0614 (1.0683) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.419, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.1232 (1.0555) acc 100.0000 (100.0000) lr 0.000260 +epoch: [303/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0533 (1.0625) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.461, TIME@all 0.312 +epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0580 (1.0590) acc 100.0000 (99.8438) lr 0.000260 +epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0815 (1.0691) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.428, TIME@all 0.312 +epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0505 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.0822 (1.0649) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.150, TIME@all 0.312 +epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:12:07 loss 1.0504 (1.0644) acc 100.0000 (99.6875) lr 0.000260 +epoch: [304/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.0865 (1.0748) acc 96.8750 (99.6875) lr 0.000260 +FPS@all 821.221, TIME@all 0.312 +epoch: [304/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0659 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0708 (1.0673) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.142, TIME@all 0.312 +epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0813 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:12:01 loss 1.0865 (1.0679) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.092, TIME@all 0.312 +epoch: [304/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:12:07 loss 1.0593 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0944 (1.0633) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.165, TIME@all 0.312 +epoch: [304/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:12:07 loss 1.0543 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0497 (1.0699) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.164, TIME@all 0.312 +epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0511 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0649 (1.0682) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.144, TIME@all 0.312 +epoch: [304/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:12:07 loss 1.0604 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [304/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.1171 (1.0640) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.188, TIME@all 0.312 +epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:11:53 loss 1.0824 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.007) eta 0:11:47 loss 1.0551 (1.0734) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.893, TIME@all 0.312 +epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0724 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.1025 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.816, TIME@all 0.312 +epoch: [305/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0863 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0717 (1.0644) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.915, TIME@all 0.312 +epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0659 (1.0594) acc 100.0000 (99.8438) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0756 (1.0672) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.832, TIME@all 0.312 +epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.011) eta 0:11:53 loss 1.0754 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0507 (1.0671) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.825, TIME@all 0.312 +epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0483 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0552 (1.0692) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.826, TIME@all 0.312 +epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:11:53 loss 1.1160 (1.0593) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.302 (0.313) data 0.000 (0.006) eta 0:11:46 loss 1.0649 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.008, TIME@all 0.312 +epoch: [305/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:11:53 loss 1.0852 (1.0623) acc 100.0000 (100.0000) lr 0.000260 +epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0530 (1.0658) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.917, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1155 (1.0596) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0760 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.137, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:11:37 loss 1.1228 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:11:30 loss 1.0664 (1.0676) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.165, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1138 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.1242 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.033, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1054 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0977 (1.0722) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.097, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1765 (1.0634) acc 96.8750 (99.8438) lr 0.000260 +epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0992 (1.0706) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.104, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1045 (1.0596) acc 100.0000 (100.0000) lr 0.000260 +epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.1213 (1.0704) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.159, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:11:37 loss 1.0818 (1.0641) acc 100.0000 (99.8438) lr 0.000260 +epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:11:30 loss 1.1011 (1.0728) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.101, TIME@all 0.312 +epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.0861 (1.0599) acc 100.0000 (99.8438) lr 0.000260 +epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0746 (1.0645) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.111, TIME@all 0.312 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0864 (1.0634) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0470 (1.0727) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.204, TIME@all 0.312 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0759 (1.0602) acc 100.0000 (99.6875) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0666 (1.0655) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.120, TIME@all 0.313 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0709 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0582 (1.0714) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.054, TIME@all 0.313 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.1140 (1.0686) acc 100.0000 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0498 (1.0701) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.042, TIME@all 0.313 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0521 (1.0616) acc 100.0000 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0553 (1.0750) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.080, TIME@all 0.313 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0597 (1.0710) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:11:15 loss 1.0552 (1.0756) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.126, TIME@all 0.313 +epoch: [307/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0657 (1.0612) acc 100.0000 (99.8438) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:11:15 loss 1.0913 (1.0693) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.098, TIME@all 0.313 +epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0647 (1.0626) acc 100.0000 (100.0000) lr 0.000260 +epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0603 (1.0726) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.077, TIME@all 0.313 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.1280 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0420 (1.0651) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.278, TIME@all 0.312 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0790 (1.0513) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0677 (1.0709) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 820.186, TIME@all 0.312 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0863 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0551 (1.0666) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.083, TIME@all 0.312 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.1026 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:11:00 loss 1.0450 (1.0636) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.126, TIME@all 0.312 +epoch: [308/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0904 (1.0577) acc 100.0000 (99.8438) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0499 (1.0621) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.183, TIME@all 0.312 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:11:08 loss 1.1121 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0491 (1.0678) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.120, TIME@all 0.312 +epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0853 (1.0529) acc 100.0000 (100.0000) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0494 (1.0652) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.094, TIME@all 0.312 +epoch: [308/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:11:08 loss 1.1473 (1.0644) acc 100.0000 (99.8438) lr 0.000260 +epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0503 (1.0691) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.145, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:49 loss 1.0724 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0588 (1.0642) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.509, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:10:50 loss 1.0706 (1.0668) acc 100.0000 (99.5312) lr 0.000260 +epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0682 (1.0702) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 821.526, TIME@all 0.312 +epoch: [309/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.1391 (1.0640) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0891 (1.0687) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.385, TIME@all 0.312 +epoch: [309/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0679 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.1503 (1.0690) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 821.369, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0596 (1.0570) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0582 (1.0652) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.415, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0512 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0585 (1.0675) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.476, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0604 (1.0597) acc 100.0000 (99.8438) lr 0.000260 +epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0712 (1.0670) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.421, TIME@all 0.312 +epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:10:50 loss 1.0495 (1.0596) acc 100.0000 (99.6875) lr 0.000260 +epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.1042 (1.0658) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.451, TIME@all 0.312 +epoch: [310/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:10:33 loss 1.0779 (1.0541) acc 100.0000 (99.8438) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:10:29 loss 1.0790 (1.0633) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.340, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0481 (1.0544) acc 100.0000 (99.8438) lr 0.000260 +epoch: [310/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0616 (1.0676) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.247, TIME@all 0.313 +epoch: [310/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0453 (1.0556) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0697 (1.0666) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.150, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:34 loss 1.0552 (1.0510) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0784 (1.0625) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.160, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:10:33 loss 1.0762 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:10:29 loss 1.0762 (1.0651) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.191, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0646 (1.0599) acc 100.0000 (99.8438) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0695 (1.0725) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 818.243, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0675 (1.0518) acc 100.0000 (100.0000) lr 0.000260 +epoch: [310/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0455 (1.0652) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.235, TIME@all 0.313 +epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0703 (1.0585) acc 100.0000 (99.8438) lr 0.000260 +epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.1379 (1.0681) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.268, TIME@all 0.313 +epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0979 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0633 (1.0660) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.719, TIME@all 0.312 +epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0604 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0522 (1.0623) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.715, TIME@all 0.312 +epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0520 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0675 (1.0607) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.549, TIME@all 0.312 +epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0626 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0691 (1.0616) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.653, TIME@all 0.312 +epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:10:19 loss 1.0681 (1.0514) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:10:13 loss 1.0478 (1.0623) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.618, TIME@all 0.312 +epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0715 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0806 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.682, TIME@all 0.312 +epoch: [311/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:10:19 loss 1.0575 (1.0550) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0665 (1.0622) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.654, TIME@all 0.312 +epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0643 (1.0513) acc 100.0000 (100.0000) lr 0.000260 +epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0641 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.660, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:10:02 loss 1.0485 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1430 (1.0677) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.100, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0513 (1.0620) acc 100.0000 (99.8438) lr 0.000260 +epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1363 (1.0683) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.010, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0462 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1134 (1.0649) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.881, TIME@all 0.312 +epoch: [312/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0466 (1.0655) acc 100.0000 (99.8438) lr 0.000260 +epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.0600 (1.0667) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.940, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0671 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1235 (1.0665) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.928, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0773 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1089 (1.0651) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.009, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0575 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:09:56 loss 1.0661 (1.0683) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.966, TIME@all 0.312 +epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0500 (1.0601) acc 100.0000 (100.0000) lr 0.000260 +epoch: [312/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1494 (1.0668) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.969, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:09:46 loss 1.0860 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0842 (1.0659) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.439, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.1177 (1.0673) acc 100.0000 (99.8438) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0673 (1.0765) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.260, TIME@all 0.312 +epoch: [313/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0515 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0461 (1.0679) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.357, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0544 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0663 (1.0708) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.254, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:09:46 loss 1.0644 (1.0635) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0719 (1.0689) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.289, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:09:46 loss 1.0646 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [313/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0703 (1.0693) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.348, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0504 (1.0548) acc 100.0000 (99.8438) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0596 (1.0680) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.346, TIME@all 0.312 +epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0630 (1.0584) acc 100.0000 (99.8438) lr 0.000260 +epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0659 (1.0683) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.333, TIME@all 0.312 +epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0636 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0679 (1.0620) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.867, TIME@all 0.312 +epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0483 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0601 (1.0650) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.901, TIME@all 0.312 +epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0682 (1.0562) acc 100.0000 (99.8438) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:09:25 loss 1.1156 (1.0727) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.799, TIME@all 0.312 +epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 0:09:31 loss 1.0514 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:09:25 loss 1.0639 (1.0623) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.791, TIME@all 0.312 +epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:09:31 loss 1.0893 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:09:25 loss 1.1218 (1.0691) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.905, TIME@all 0.312 +epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0627 (1.0498) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0810 (1.0596) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.809, TIME@all 0.312 +epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0515 (1.0520) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 0:09:25 loss 1.0493 (1.0665) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.850, TIME@all 0.312 +epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0516 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:09:25 loss 1.0838 (1.0641) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.853, TIME@all 0.312 +epoch: [315/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0772 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0742 (1.0726) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.839, TIME@all 0.312 +epoch: [315/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0878 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0831 (1.0682) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.704, TIME@all 0.312 +epoch: [315/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0829 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1058 (1.0693) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.736, TIME@all 0.312 +epoch: [315/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0809 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1148 (1.0683) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 820.756, TIME@all 0.312 +epoch: [315/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.1072 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0846 (1.0648) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.781, TIME@all 0.312 +epoch: [315/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0903 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0961 (1.0686) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.754, TIME@all 0.312 +epoch: [315/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0733 (1.0618) acc 100.0000 (99.8438) lr 0.000260 +epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1059 (1.0675) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.770, TIME@all 0.312 +epoch: [315/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0968 (1.0587) acc 100.0000 (100.0000) lr 0.000260 +epoch: [315/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:09:09 loss 1.0558 (1.0643) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.731, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0584 (1.0541) acc 100.0000 (99.8438) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0668 (1.0657) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.112, TIME@all 0.312 +epoch: [316/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0457 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0883 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.987, TIME@all 0.312 +epoch: [316/350][20/50] time 0.318 (0.313) data 0.000 (0.011) eta 0:09:00 loss 1.0502 (1.0547) acc 100.0000 (99.8438) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0728 (1.0677) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.939, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0558 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0515 (1.0652) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.959, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0478 (1.0609) acc 100.0000 (99.8438) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0542 (1.0673) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.978, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0498 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0813 (1.0634) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.997, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0530 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:08:54 loss 1.0561 (1.0652) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.989, TIME@all 0.312 +epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0703 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [316/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0559 (1.0632) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.961, TIME@all 0.312 +epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:08:44 loss 1.0597 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.1243 (1.0699) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.947, TIME@all 0.311 +epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0520 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0640 (1.0704) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.890, TIME@all 0.311 +epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0557 (1.0674) acc 100.0000 (99.6875) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0852 (1.0742) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 821.803, TIME@all 0.312 +epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0793 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0482 (1.0656) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.816, TIME@all 0.312 +epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0589 (1.0631) acc 100.0000 (99.8438) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0560 (1.0685) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.839, TIME@all 0.311 +epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0551 (1.0654) acc 100.0000 (99.5312) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0795 (1.0732) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 821.870, TIME@all 0.311 +epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0610 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0693 (1.0673) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.863, TIME@all 0.311 +epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:08:44 loss 1.0489 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0641 (1.0631) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.810, TIME@all 0.312 +epoch: [318/350][20/50] time 0.315 (0.312) data 0.000 (0.014) eta 0:08:29 loss 1.0519 (1.0521) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:08:23 loss 1.0562 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.076, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0893 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0594 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.914, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:08:28 loss 1.0715 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:08:24 loss 1.0655 (1.0662) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.951, TIME@all 0.313 +epoch: [318/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0675 (1.0571) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0519 (1.0681) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.001, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0527 (1.0600) acc 100.0000 (99.8438) lr 0.000260 +epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0499 (1.0688) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.998, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0673 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:08:24 loss 1.0642 (1.0653) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.993, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0658 (1.0625) acc 100.0000 (99.8438) lr 0.000260 +epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0660 (1.0689) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.948, TIME@all 0.313 +epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0908 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0686 (1.0629) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.970, TIME@all 0.313 +epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:08:13 loss 1.0868 (1.0626) acc 100.0000 (99.6875) lr 0.000260 +epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:06 loss 1.0413 (1.0629) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.607, TIME@all 0.312 +epoch: [319/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:13 loss 1.1006 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:06 loss 1.1227 (1.0642) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.666, TIME@all 0.312 +epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.1053 (1.0622) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0565 (1.0641) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.512, TIME@all 0.312 +epoch: [319/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0490 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0758 (1.0616) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.508, TIME@all 0.312 +epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0946 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:08:06 loss 1.0506 (1.0592) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.621, TIME@all 0.312 +epoch: [319/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:08:13 loss 1.0734 (1.0563) acc 100.0000 (99.8438) lr 0.000260 +epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:08:07 loss 1.0532 (1.0624) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.534, TIME@all 0.312 +epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0852 (1.0605) acc 100.0000 (99.8438) lr 0.000260 +epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0568 (1.0645) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.550, TIME@all 0.312 +epoch: [319/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:08:13 loss 1.0931 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:07 loss 1.1334 (1.0636) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 821.572, TIME@all 0.312 +epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.1170 (1.0590) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0602 (1.0653) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.284, TIME@all 0.312 +epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0842 (1.0620) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0567 (1.0720) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.190, TIME@all 0.312 +epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1147 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0675 (1.0712) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.270, TIME@all 0.312 +epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0756 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0463 (1.0619) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.189, TIME@all 0.312 +epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0744 (1.0558) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0827 (1.0652) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.264, TIME@all 0.312 +epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0834 (1.0601) acc 100.0000 (99.8438) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0526 (1.0654) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.212, TIME@all 0.312 +epoch: [320/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1019 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0490 (1.0626) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.227, TIME@all 0.312 +epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1420 (1.0606) acc 100.0000 (100.0000) lr 0.000260 +epoch: [320/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0470 (1.0662) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.229, TIME@all 0.312 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0854 (1.0625) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0983 (1.0662) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 816.530, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.1093 (1.0596) acc 100.0000 (99.8438) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:38 loss 1.1638 (1.0695) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 816.554, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.0617 (1.0621) acc 100.0000 (99.6875) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.1430 (1.0706) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 816.384, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.0615 (1.0588) acc 100.0000 (99.6875) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.0731 (1.0624) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 816.390, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.1208 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.1226 (1.0667) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 816.429, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0975 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0956 (1.0642) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 816.449, TIME@all 0.314 +epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.1480 (1.0621) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.0572 (1.0644) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 816.398, TIME@all 0.314 +epoch: [321/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0600 (1.0571) acc 100.0000 (100.0000) lr 0.000260 +epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0844 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 816.438, TIME@all 0.314 +epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0448 (1.0662) acc 100.0000 (99.5312) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1174 (1.0738) acc 100.0000 (99.5312) lr 0.000260 +FPS@all 821.680, TIME@all 0.312 +epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0871 (1.0613) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1242 (1.0696) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.687, TIME@all 0.312 +epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:07:25 loss 1.0671 (1.0583) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0889 (1.0670) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.542, TIME@all 0.312 +epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:07:25 loss 1.0527 (1.0589) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0902 (1.0638) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.533, TIME@all 0.312 +epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0508 (1.0627) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0740 (1.0680) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.596, TIME@all 0.312 +epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0543 (1.0581) acc 100.0000 (100.0000) lr 0.000260 +epoch: [322/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1159 (1.0698) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.628, TIME@all 0.312 +epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0607 (1.0628) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0796 (1.0675) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.591, TIME@all 0.312 +epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0733 (1.0571) acc 100.0000 (99.8438) lr 0.000260 +epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1503 (1.0697) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 821.546, TIME@all 0.312 +epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0611 (1.0527) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0829 (1.0646) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.789, TIME@all 0.313 +epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0518 (1.0567) acc 100.0000 (99.8438) lr 0.000260 +epoch: [323/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0552 (1.0660) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.837, TIME@all 0.313 +epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0500 (1.0531) acc 100.0000 (99.8438) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0974 (1.0629) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.760, TIME@all 0.313 +epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:11 loss 1.0471 (1.0507) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:07:05 loss 1.0873 (1.0650) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.689, TIME@all 0.313 +epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0522 (1.0547) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.1042 (1.0711) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 818.723, TIME@all 0.313 +epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0554 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0987 (1.0635) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.753, TIME@all 0.313 +epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0510 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0716 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.763, TIME@all 0.313 +epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0539 (1.0526) acc 100.0000 (100.0000) lr 0.000260 +epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0941 (1.0694) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.813, TIME@all 0.313 +epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0604 (1.0663) acc 100.0000 (99.6875) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0680 (1.0680) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.543, TIME@all 0.312 +epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:56 loss 1.0995 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:06:49 loss 1.1361 (1.0661) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 819.591, TIME@all 0.312 +epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0583 (1.0523) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0467 (1.0574) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.438, TIME@all 0.312 +epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0448 (1.0602) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0475 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.461, TIME@all 0.312 +epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0492 (1.0494) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:06:50 loss 1.0465 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.558, TIME@all 0.312 +epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0553 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:06:50 loss 1.0670 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.495, TIME@all 0.312 +epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0513 (1.0523) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0610 (1.0613) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.499, TIME@all 0.312 +epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:06:56 loss 1.0779 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [324/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:06:50 loss 1.0609 (1.0640) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.481, TIME@all 0.312 +epoch: [325/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0610 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1183 (1.0682) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.033, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0561 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.1638 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.924, TIME@all 0.312 +epoch: [325/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0597 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:06:33 loss 1.1179 (1.0667) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.825, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0674 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.0729 (1.0621) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.845, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0554 (1.0562) acc 100.0000 (99.8438) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.1291 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.889, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0776 (1.0619) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1381 (1.0696) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.860, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0615 (1.0601) acc 100.0000 (100.0000) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1161 (1.0660) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 820.916, TIME@all 0.312 +epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0510 (1.0635) acc 100.0000 (99.6875) lr 0.000260 +epoch: [325/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 0:06:33 loss 1.0934 (1.0702) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.923, TIME@all 0.312 +epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0458 (1.0557) acc 100.0000 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0561 (1.0667) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.178, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:06:25 loss 1.0541 (1.0600) acc 100.0000 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0544 (1.0726) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.057, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0546 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0634 (1.0613) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.018, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0544 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0444 (1.0614) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.059, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0626 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0579 (1.0626) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.099, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0554 (1.0537) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.1215 (1.0656) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 819.094, TIME@all 0.313 +epoch: [326/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0572 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0893 (1.0638) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.118, TIME@all 0.313 +epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:06:25 loss 1.0700 (1.0567) acc 96.8750 (99.8438) lr 0.000260 +epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0470 (1.0629) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.058, TIME@all 0.313 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0589 (1.0570) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:06:02 loss 1.0690 (1.0664) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.360, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0705 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0547 (1.0645) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.323, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0699 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.1064 (1.0647) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.233, TIME@all 0.312 +epoch: [327/350][20/50] time 0.308 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0566 (1.0514) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0658 (1.0635) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.208, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.1057 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0533 (1.0690) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.316, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:06:09 loss 1.0597 (1.0560) acc 100.0000 (99.8438) lr 0.000260 +epoch: [327/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.1235 (1.0688) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.276, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:06:09 loss 1.0656 (1.0578) acc 100.0000 (99.8438) lr 0.000260 +epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:02 loss 1.0623 (1.0645) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.273, TIME@all 0.312 +epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0651 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [327/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:06:02 loss 1.0673 (1.0695) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.290, TIME@all 0.312 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:05:52 loss 1.1806 (1.0669) acc 96.8750 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:45 loss 1.0713 (1.0734) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 822.428, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.1453 (1.0626) acc 96.8750 (99.6875) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.1033 (1.0695) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.355, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.1064 (1.0624) acc 96.8750 (99.6875) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.1597 (1.0768) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 822.256, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0622 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0946 (1.0683) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 822.244, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0704 (1.0591) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0963 (1.0717) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 822.323, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0818 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0842 (1.0701) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 822.281, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:05:52 loss 1.1262 (1.0642) acc 100.0000 (100.0000) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:46 loss 1.0842 (1.0732) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.265, TIME@all 0.311 +epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0647 (1.0594) acc 100.0000 (99.8438) lr 0.000260 +epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0976 (1.0763) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 822.301, TIME@all 0.311 +epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0639 (1.0536) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0512 (1.0629) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.017, TIME@all 0.312 +epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0660 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0536 (1.0639) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.076, TIME@all 0.312 +epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:37 loss 1.0541 (1.0517) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:05:31 loss 1.1735 (1.0650) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 819.966, TIME@all 0.312 +epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0909 (1.0609) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0493 (1.0655) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.997, TIME@all 0.312 +epoch: [329/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0819 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:05:31 loss 1.0581 (1.0651) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.928, TIME@all 0.312 +epoch: [329/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0839 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0811 (1.0692) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.951, TIME@all 0.312 +epoch: [329/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:05:37 loss 1.0694 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:05:31 loss 1.0641 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.010, TIME@all 0.312 +epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0844 (1.0729) acc 100.0000 (99.6875) lr 0.000260 +epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0590 (1.0731) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.960, TIME@all 0.312 +epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0690 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0577 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 822.983, TIME@all 0.311 +epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0456 (1.0552) acc 100.0000 (99.8438) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0812 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 823.021, TIME@all 0.311 +epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0552 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:05:14 loss 1.0527 (1.0642) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.834, TIME@all 0.311 +epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0527 (1.0606) acc 100.0000 (99.6875) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0490 (1.0662) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.915, TIME@all 0.311 +epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.012) eta 0:05:20 loss 1.0518 (1.0652) acc 100.0000 (99.6875) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:05:14 loss 1.0466 (1.0740) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.871, TIME@all 0.311 +epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0708 (1.0582) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0711 (1.0650) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 822.925, TIME@all 0.311 +epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0640 (1.0571) acc 100.0000 (99.8438) lr 0.000260 +epoch: [330/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:14 loss 1.0490 (1.0661) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 822.855, TIME@all 0.311 +epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0885 (1.0590) acc 100.0000 (100.0000) lr 0.000260 +epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0502 (1.0647) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 822.911, TIME@all 0.311 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:06 loss 1.0901 (1.0592) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:05:00 loss 1.0593 (1.0689) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.835, TIME@all 0.312 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0799 (1.0572) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0614 (1.0716) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.621, TIME@all 0.312 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0605 (1.0625) acc 100.0000 (99.8438) lr 0.000260 +epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0643 (1.0702) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.605, TIME@all 0.312 +epoch: [331/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0677 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0596 (1.0654) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.682, TIME@all 0.312 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0611 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0695 (1.0672) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.669, TIME@all 0.312 +epoch: [331/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0610 (1.0577) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0794 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.704, TIME@all 0.312 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0635 (1.0594) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0559 (1.0680) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.691, TIME@all 0.312 +epoch: [331/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:05:06 loss 1.0577 (1.0593) acc 100.0000 (100.0000) lr 0.000260 +epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0797 (1.0681) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.742, TIME@all 0.312 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:52 loss 1.0767 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0561 (1.0667) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 817.682, TIME@all 0.313 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:52 loss 1.0670 (1.0560) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0951 (1.0703) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 817.681, TIME@all 0.313 +epoch: [332/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:04:53 loss 1.0755 (1.0585) acc 100.0000 (99.8438) lr 0.000260 +epoch: [332/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0602 (1.0713) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 817.509, TIME@all 0.313 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:04:53 loss 1.1067 (1.0590) acc 96.8750 (99.6875) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0651 (1.0731) acc 100.0000 (99.6094) lr 0.000260 +FPS@all 817.530, TIME@all 0.313 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.001 (0.013) eta 0:04:53 loss 1.0764 (1.0609) acc 100.0000 (99.6875) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0436 (1.0721) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 817.567, TIME@all 0.313 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.001 (0.014) eta 0:04:53 loss 1.0564 (1.0550) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.001 (0.007) eta 0:04:45 loss 1.0537 (1.0659) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 817.609, TIME@all 0.313 +epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:53 loss 1.0473 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0729 (1.0685) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 817.572, TIME@all 0.313 +epoch: [332/350][20/50] time 0.313 (0.315) data 0.001 (0.013) eta 0:04:52 loss 1.0546 (1.0581) acc 100.0000 (100.0000) lr 0.000260 +epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0479 (1.0749) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 817.660, TIME@all 0.313 +epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0808 (1.0644) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0820 (1.0628) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.462, TIME@all 0.312 +epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0444 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0738 (1.0634) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.356, TIME@all 0.312 +epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0981 (1.0629) acc 100.0000 (99.8438) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0474 (1.0640) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.293, TIME@all 0.312 +epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0506 (1.0603) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0610 (1.0637) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.277, TIME@all 0.312 +epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0710 (1.0593) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0719 (1.0668) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.331, TIME@all 0.312 +epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0648 (1.0600) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0962 (1.0667) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.306, TIME@all 0.312 +epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0706 (1.0545) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0739 (1.0621) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.306, TIME@all 0.312 +epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0720 (1.0563) acc 100.0000 (100.0000) lr 0.000260 +epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0611 (1.0597) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.339, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.015) eta 0:04:18 loss 1.0576 (1.0613) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0926 (1.0671) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.447, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0466 (1.0629) acc 100.0000 (99.8438) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0799 (1.0666) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.303, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0460 (1.0664) acc 100.0000 (99.8438) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1131 (1.0696) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.385, TIME@all 0.312 +epoch: [334/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0556 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:04:12 loss 1.0693 (1.0684) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.346, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0471 (1.0575) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1053 (1.0660) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.299, TIME@all 0.312 +epoch: [334/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0460 (1.0531) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1003 (1.0647) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.407, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:04:18 loss 1.0660 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0776 (1.0688) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.418, TIME@all 0.312 +epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0558 (1.0617) acc 100.0000 (99.8438) lr 0.000260 +epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0774 (1.0659) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.402, TIME@all 0.312 +epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0738 (1.0588) acc 100.0000 (99.8438) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0834 (1.0678) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.776, TIME@all 0.313 +epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0673 (1.0669) acc 100.0000 (99.5312) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.1098 (1.0707) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 818.806, TIME@all 0.313 +epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0981 (1.0584) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:03:58 loss 1.0846 (1.0683) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.696, TIME@all 0.313 +epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0697 (1.0614) acc 100.0000 (99.8438) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0875 (1.0659) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.781, TIME@all 0.313 +epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0612 (1.0542) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0718 (1.0622) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.601, TIME@all 0.313 +epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0800 (1.0572) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:03:58 loss 1.0933 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.672, TIME@all 0.313 +epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:04:04 loss 1.0637 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:03:58 loss 1.0974 (1.0626) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.660, TIME@all 0.313 +epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0492 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0725 (1.0663) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.720, TIME@all 0.313 +epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0675 (1.0554) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0862 (1.0644) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.947, TIME@all 0.312 +epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 0:03:48 loss 1.1224 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:41 loss 1.1045 (1.0634) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.011, TIME@all 0.312 +epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:03:48 loss 1.0631 (1.0537) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:03:42 loss 1.0745 (1.0646) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.877, TIME@all 0.312 +epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0647 (1.0556) acc 100.0000 (99.8438) lr 0.000260 +epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.1083 (1.0696) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 819.906, TIME@all 0.312 +epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0872 (1.0611) acc 100.0000 (99.8438) lr 0.000260 +epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:03:42 loss 1.2024 (1.0672) acc 93.7500 (99.6094) lr 0.000260 +FPS@all 819.902, TIME@all 0.312 +epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0578 (1.0515) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0598 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.954, TIME@all 0.312 +epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0572 (1.0512) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0502 (1.0617) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.935, TIME@all 0.312 +epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.1493 (1.0569) acc 100.0000 (100.0000) lr 0.000260 +epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0642 (1.0606) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.914, TIME@all 0.312 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.014) eta 0:03:32 loss 1.0526 (1.0596) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:03:26 loss 1.1048 (1.0750) acc 96.8750 (99.6875) lr 0.000260 +FPS@all 818.500, TIME@all 0.313 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:33 loss 1.0581 (1.0597) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0587 (1.0698) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 818.280, TIME@all 0.313 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0566 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0645 (1.0647) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.366, TIME@all 0.313 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0682 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0883 (1.0675) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.393, TIME@all 0.313 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0616 (1.0612) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0630 (1.0659) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.356, TIME@all 0.313 +epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0615 (1.0643) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0774 (1.0688) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.332, TIME@all 0.313 +epoch: [337/350][20/50] time 0.320 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0844 (1.0719) acc 100.0000 (99.8438) lr 0.000260 +epoch: [337/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:03:26 loss 1.0720 (1.0728) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.411, TIME@all 0.313 +epoch: [337/350][20/50] time 0.320 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0528 (1.0613) acc 100.0000 (100.0000) lr 0.000260 +epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0717 (1.0717) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 818.377, TIME@all 0.313 +epoch: [338/350][20/50] time 0.306 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.0634 (1.0590) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0681 (1.0675) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.059, TIME@all 0.312 +epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.014) eta 0:03:16 loss 1.0836 (1.0588) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0463 (1.0626) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.076, TIME@all 0.312 +epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.0915 (1.0621) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0640 (1.0672) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.970, TIME@all 0.312 +epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.1081 (1.0584) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0595 (1.0653) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.965, TIME@all 0.312 +epoch: [338/350][20/50] time 0.306 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.1365 (1.0584) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0602 (1.0670) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 821.017, TIME@all 0.312 +epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.1009 (1.0557) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0574 (1.0611) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.023, TIME@all 0.312 +epoch: [338/350][20/50] time 0.307 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.0650 (1.0558) acc 100.0000 (99.8438) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0723 (1.0613) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.038, TIME@all 0.312 +epoch: [338/350][20/50] time 0.306 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.0768 (1.0562) acc 100.0000 (100.0000) lr 0.000260 +epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0490 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.057, TIME@all 0.312 +epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0695 (1.0540) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0503 (1.0667) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.803, TIME@all 0.312 +epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:03:01 loss 1.0574 (1.0586) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:02:55 loss 1.0478 (1.0653) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.829, TIME@all 0.312 +epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0536 (1.0552) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0509 (1.0687) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.708, TIME@all 0.312 +epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0436 (1.0520) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0496 (1.0667) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.695, TIME@all 0.312 +epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0506 (1.0496) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0740 (1.0582) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.779, TIME@all 0.312 +epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:03:01 loss 1.0737 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:02:55 loss 1.0481 (1.0637) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.770, TIME@all 0.312 +epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0549 (1.0555) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0655 (1.0649) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.727, TIME@all 0.312 +epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0588 (1.0549) acc 100.0000 (99.8438) lr 0.000260 +epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0989 (1.0604) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.756, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0911 (1.0619) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0847 (1.0669) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.889, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0565 (1.0569) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0529 (1.0607) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.886, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.0528 (1.0549) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0468 (1.0586) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.778, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.314 (0.312) data 0.000 (0.011) eta 0:02:45 loss 1.0969 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0581 (1.0606) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.747, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.0611 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0485 (1.0566) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.817, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.1009 (1.0590) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.1056 (1.0669) acc 96.8750 (99.8438) lr 0.000260 +FPS@all 820.815, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0856 (1.0568) acc 100.0000 (99.8438) lr 0.000260 +epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0842 (1.0674) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.790, TIME@all 0.312 +Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 +epoch: [340/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.1098 (1.0608) acc 100.0000 (100.0000) lr 0.000260 +epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0582 (1.0648) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.790, TIME@all 0.312 +epoch: [341/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.0687 (1.0564) acc 100.0000 (99.8438) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0511 (1.0651) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 820.330, TIME@all 0.312 +epoch: [341/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.0648 (1.0599) acc 100.0000 (99.8438) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0552 (1.0670) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.375, TIME@all 0.312 +epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.0765 (1.0598) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:02:23 loss 1.1107 (1.0676) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.243, TIME@all 0.312 +epoch: [341/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:02:30 loss 1.0735 (1.0574) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:02:23 loss 1.1124 (1.0686) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.244, TIME@all 0.312 +epoch: [341/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.1659 (1.0639) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0693 (1.0720) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.264, TIME@all 0.312 +epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.014) eta 0:02:30 loss 1.0822 (1.0605) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:02:23 loss 1.1309 (1.0720) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.264, TIME@all 0.312 +epoch: [341/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.1143 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:02:23 loss 1.1079 (1.0650) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.352, TIME@all 0.312 +epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.0578 (1.0539) acc 100.0000 (100.0000) lr 0.000260 +epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:02:23 loss 1.0594 (1.0665) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.262, TIME@all 0.312 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0487 (1.0595) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0773 (1.0668) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 817.959, TIME@all 0.313 +epoch: [342/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0526 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0959 (1.0638) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 817.874, TIME@all 0.313 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:02:14 loss 1.0464 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:02:08 loss 1.1071 (1.0742) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 818.011, TIME@all 0.313 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0498 (1.0553) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.1374 (1.0679) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 817.886, TIME@all 0.313 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0598 (1.0528) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0675 (1.0671) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 817.976, TIME@all 0.313 +epoch: [342/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:02:14 loss 1.0878 (1.0567) acc 96.8750 (99.8438) lr 0.000260 +epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:02:08 loss 1.1174 (1.0701) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 817.958, TIME@all 0.313 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:02:14 loss 1.0700 (1.0610) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0915 (1.0741) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 817.880, TIME@all 0.313 +epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0520 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0741 (1.0664) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 817.922, TIME@all 0.313 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1790 (1.0651) acc 96.8750 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0628 (1.0657) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.257, TIME@all 0.312 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1094 (1.0618) acc 100.0000 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0718 (1.0662) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.236, TIME@all 0.312 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.0799 (1.0599) acc 100.0000 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0724 (1.0641) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.142, TIME@all 0.313 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1809 (1.0588) acc 96.8750 (99.8438) lr 0.000260 +epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0603 (1.0635) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.144, TIME@all 0.313 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:01:59 loss 1.1862 (1.0580) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:01:52 loss 1.0564 (1.0684) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.087, TIME@all 0.313 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1345 (1.0550) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0522 (1.0661) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.214, TIME@all 0.312 +epoch: [343/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1598 (1.0633) acc 100.0000 (100.0000) lr 0.000260 +epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0675 (1.0717) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.151, TIME@all 0.313 +epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.2321 (1.0618) acc 93.7500 (99.6875) lr 0.000260 +epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0636 (1.0682) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 819.145, TIME@all 0.313 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:01:42 loss 1.0571 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0518 (1.0586) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.193, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0577 (1.0532) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0450 (1.0577) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.087, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0762 (1.0499) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0864 (1.0638) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.040, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0481 (1.0535) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0508 (1.0603) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.111, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 0:01:42 loss 1.0566 (1.0555) acc 100.0000 (99.8438) lr 0.000260 +epoch: [344/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:01:36 loss 1.0479 (1.0633) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.077, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0702 (1.0506) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0544 (1.0640) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.041, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0508 (1.0510) acc 100.0000 (99.8438) lr 0.000260 +epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0637 (1.0595) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.104, TIME@all 0.312 +epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0489 (1.0536) acc 100.0000 (100.0000) lr 0.000260 +epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0778 (1.0649) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.109, TIME@all 0.312 +epoch: [345/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0945 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.319 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0594 (1.0617) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.408, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0892 (1.0644) acc 100.0000 (99.8438) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:01:21 loss 1.0581 (1.0691) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.269, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0969 (1.0561) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0672 (1.0636) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.347, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:01:27 loss 1.0729 (1.0535) acc 100.0000 (99.8438) lr 0.000260 +epoch: [345/350][40/50] time 0.319 (0.312) data 0.000 (0.006) eta 0:01:21 loss 1.0493 (1.0609) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.274, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.1895 (1.0618) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.001 (0.007) eta 0:01:21 loss 1.0500 (1.0695) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.384, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.2562 (1.0651) acc 96.8750 (99.8438) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0717 (1.0693) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.317, TIME@all 0.312 +epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0719 (1.0557) acc 100.0000 (99.8438) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0494 (1.0636) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.303, TIME@all 0.312 +epoch: [345/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0835 (1.0543) acc 100.0000 (100.0000) lr 0.000260 +epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0898 (1.0644) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.291, TIME@all 0.312 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0497 (1.0546) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1122 (1.0617) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.456, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0555 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1038 (1.0626) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.407, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0624 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:01:05 loss 1.1729 (1.0616) acc 96.8750 (99.9219) lr 0.000260 +FPS@all 818.507, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0662 (1.0521) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1349 (1.0619) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 818.407, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0744 (1.0579) acc 100.0000 (99.8438) lr 0.000260 +epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1691 (1.0664) acc 96.8750 (99.6875) lr 0.000260 +FPS@all 818.427, TIME@all 0.313 +epoch: [346/350][20/50] time 0.325 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0577 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.0776 (1.0586) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.470, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0602 (1.0514) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:01:05 loss 1.0892 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.432, TIME@all 0.313 +epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0562 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.0712 (1.0609) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 818.443, TIME@all 0.313 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0501 (1.0604) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0852 (1.0771) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 819.651, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0512 (1.0530) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0831 (1.0665) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.729, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0468 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0587 (1.0660) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.535, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:00:56 loss 1.0526 (1.0564) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0668 (1.0681) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.577, TIME@all 0.312 +epoch: [347/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0584 (1.0544) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0684 (1.0679) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.619, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:56 loss 1.0508 (1.0565) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.1010 (1.0696) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.610, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0490 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0832 (1.0663) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.600, TIME@all 0.312 +epoch: [347/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:56 loss 1.0497 (1.0541) acc 100.0000 (100.0000) lr 0.000260 +epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0699 (1.0660) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.703, TIME@all 0.312 +epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.014) eta 0:00:40 loss 1.1071 (1.0551) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0751 (1.0655) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.177, TIME@all 0.313 +epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.0941 (1.0540) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0486 (1.0642) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 819.179, TIME@all 0.313 +epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1209 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0496 (1.0656) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.064, TIME@all 0.313 +epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:00:40 loss 1.2092 (1.0601) acc 96.8750 (99.8438) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:00:34 loss 1.0625 (1.0714) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.071, TIME@all 0.313 +epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1372 (1.0558) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.1040 (1.0651) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.151, TIME@all 0.313 +epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1040 (1.0513) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0579 (1.0619) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.155, TIME@all 0.313 +epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1470 (1.0585) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0736 (1.0645) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 819.121, TIME@all 0.313 +epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1153 (1.0579) acc 100.0000 (100.0000) lr 0.000260 +epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0554 (1.0665) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 819.111, TIME@all 0.313 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0508 (1.0583) acc 100.0000 (99.8438) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0724 (1.0641) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.596, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0682 (1.0525) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:00:18 loss 1.0900 (1.0678) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 820.516, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:00:24 loss 1.0553 (1.0566) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:00:18 loss 1.0561 (1.0629) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.522, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0553 (1.0573) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0709 (1.0694) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.641, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0815 (1.0576) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0537 (1.0666) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.580, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0457 (1.0597) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0429 (1.0667) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.530, TIME@all 0.312 +epoch: [349/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0662 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0616 (1.0621) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.578, TIME@all 0.312 +epoch: [349/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 0:00:24 loss 1.0489 (1.0568) acc 100.0000 (100.0000) lr 0.000260 +epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0460 (1.0665) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.626, TIME@all 0.312 +epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0722 (1.0533) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0573 (1.0664) acc 100.0000 (99.9219) lr 0.000260 +FPS@all 821.065, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0715 (1.0545) acc 100.0000 (99.8438) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0659 (1.0664) acc 100.0000 (99.7656) lr 0.000260 +FPS@all 821.109, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0912 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0591 (1.0624) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 820.978, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:00:09 loss 1.0810 (1.0548) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:00:03 loss 1.0499 (1.0673) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.976, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0627 (1.0663) acc 100.0000 (99.8438) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0509 (1.0708) acc 100.0000 (99.6875) lr 0.000260 +FPS@all 821.015, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0629 (1.0567) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0840 (1.0677) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.036, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.1065 (1.0538) acc 100.0000 (100.0000) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 0:00:03 loss 1.0758 (1.0587) acc 100.0000 (100.0000) lr 0.000260 +FPS@all 821.022, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0700 (1.0588) acc 100.0000 (99.8438) lr 0.000260 +epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0651 (1.0657) acc 100.0000 (99.8438) lr 0.000260 +FPS@all 820.982, TIME@all 0.312 +=> Final test +##### Evaluating market1501 (source) ##### +Extracting features from query set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 3368-by-512 matrix +Extracting features from gallery set ... +Done, obtained 15913-by-512 matrix +Speed: 0.0306 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.5% +CMC curve +Rank-1 : 92.3% +Rank-5 : 97.1% +Rank-10 : 98.1% +Rank-20 : 98.8% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:25 +FPS@all 819.137, TIME@all 0.313 +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0296 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.6% +CMC curve +Rank-1 : 92.3% +Rank-5 : 97.2% +Rank-10 : 98.2% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:33 +FPS@all 819.169, TIME@all 0.313 +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0321 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.6% +CMC curve +Rank-1 : 92.3% +Rank-5 : 97.2% +Rank-10 : 98.1% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:38 +FPS@all 819.180, TIME@all 0.313 +Done, obtained 15913-by-512 matrix +Speed: 0.0315 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.5% +CMC curve +Rank-1 : 92.1% +Rank-5 : 97.0% +Rank-10 : 98.1% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:38 +FPS@all 819.124, TIME@all 0.313 +Done, obtained 15913-by-512 matrix +Speed: 0.0304 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.5% +CMC curve +Rank-1 : 92.2% +Rank-5 : 97.2% +Rank-10 : 98.2% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:41 +FPS@all 819.160, TIME@all 0.313 +Done, obtained 15913-by-512 matrix +Speed: 0.0357 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.5% +CMC curve +Rank-1 : 92.3% +Rank-5 : 97.2% +Rank-10 : 98.2% +Rank-20 : 98.8% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:41 +FPS@all 819.219, TIME@all 0.312 +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +Done, obtained 15913-by-512 matrix +Speed: 0.0328 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.6% +CMC curve +Rank-1 : 92.2% +Rank-5 : 97.2% +Rank-10 : 98.2% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:42 +FPS@all 819.277, TIME@all 0.312 +Done, obtained 15913-by-512 matrix +Speed: 0.0315 sec/batch +Computing distance matrix with metric=euclidean ... +Computing CMC and mAP ... +** Results ** +mAP: 79.5% +CMC curve +Rank-1 : 92.3% +Rank-5 : 97.1% +Rank-10 : 98.0% +Rank-20 : 98.9% +Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" +Elapsed 1:47:44 +FPS@all 819.199, TIME@all 0.313 +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. +[INFO] Float status is overflow! +[INFO] Float status is overflow! +[INFO] Float status is overflow! +THPModule_npu_shutdown success. -- Gitee From 7939acdde0fb08440e233dd3b338405506f2e2c3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:47:45 +0000 Subject: [PATCH 29/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/nohup.out?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../contrib/cv/classification/OSNet/nohup.out | 22941 ---------------- 1 file changed, 22941 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/nohup.out diff --git a/PyTorch/contrib/cv/classification/OSNet/nohup.out b/PyTorch/contrib/cv/classification/OSNet/nohup.out deleted file mode 100644 index 0491b2f677..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/nohup.out +++ /dev/null @@ -1,22941 +0,0 @@ -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 0 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Traceback (most recent call last): - File "main.py", line 310, in - main() - File "main.py", line 243, in main - torch.distributed.init_process_group(backend='hccl', rank=args.local_rank, world_size=args.device_num) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 420, in init_process_group - store, rank, world_size = next(rendezvous_iterator) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/distributed/rendezvous.py", line 172, in _env_rendezvous_handler - store = TCPStore(master_addr, master_port, world_size, start_daemon, timeout) -RuntimeError: Address already in use -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 6 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 7 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 0 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 1 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 4 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 5 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 2 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 3 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 6 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 7 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 1 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 3 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 4 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 5 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 2 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 1 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.5.0+ascend.post3 -Is debug build: No -CUDA used to build PyTorch: None - -OS: CentOS Linux 7 (AltArch) -GCC version: (GCC) 7.3.0 -CMake version: version 3.18.6 - -Python version: 3.7 -Is CUDA available: No -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.5.0+ascend.post3.20210930 -[conda] torch 1.5.0+ascend.post3.20210930 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -/home/zbk/OSNet/torchreid/metrics/rank.py:59: UserWarning: Cython evaluation (very fast so highly recommended) is unavailable, now use python evaluation. - 'Cython evaluation (very fast so highly recommended) is ' -/home/zbk/OSNet/torchreid/data/datasets/image/market1501.py:83: UserWarning: The current data structure is deprecated. Please put data folders such as "bounding_box_train" under "Market-1501-v15.09.15". - 'The current data structure is deprecated. Please ' -Use npu fused optimizer -Use npu fused optimizer -Use npu fused optimizer -Use npu fused optimizer -Use npu fused optimizer -Use npu fused optimizer -Use npu fused optimizer -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -THPModule_npu_shutdown success. -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -THPModule_npu_shutdown success. -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -THPModule_npu_shutdown success. -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -Traceback (most recent call last): - File "main.py", line 310, in - if __name__ == '__main__': - File "main.py", line 286, in main - find_unused_parameters=True, - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 307, in __init__ - self.broadcast_bucket_size) - File "/root/archiconda3/envs/ych/lib/python3.7/site-packages/torch/nn/parallel/distributed.py", line 506, in _distributed_broadcast_coalesced - dist._broadcast_coalesced(self.process_group, tensors, buffer_size) -RuntimeError: HCCL error in: /usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/../c10d/HCCLUtils.hpp53, 7 -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. - -Defaults for this optimization level are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : dynamic -combine_grad : None -check_combined_tensors : None -Processing user overrides (additional kwargs that are not None)... -After processing overrides, optimization options are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : 128.0 -combine_grad : True -check_combined_tensors : None -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -group num: 1 -epoch: [1/1][20/50] time 0.286 (0.294) data 0.000 (0.020) eta 0:00:08 loss 6.5586 (6.5816) acc 0.0000 (0.9375) lr 0.260000 -epoch: [1/1][40/50] time 0.285 (0.295) data 0.000 (0.010) eta 0:00:02 loss 6.0959 (6.4039) acc 0.0000 (1.0156) lr 0.260000 -FPS@all 866.618, TIME@all 0.295 -Elapsed 0:07:53 -FPS@all 866.618, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.285 (0.294) data 0.000 (0.015) eta 1:25:43 loss 6.6704 (6.6383) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 6.1446 (6.4308) acc 3.1250 (0.7812) lr 0.260000 -FPS@all 866.705, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.286 (0.294) data 0.000 (0.017) eta 1:25:42 loss 6.6813 (6.5624) acc 0.0000 (0.7812) lr 0.260000 -epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.009) eta 1:25:48 loss 5.7973 (6.3714) acc 3.1250 (1.3281) lr 0.260000 -FPS@all 866.715, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.284 (0.295) data 0.000 (0.016) eta 1:25:53 loss 6.2679 (6.5310) acc 6.2500 (1.4062) lr 0.260000 -epoch: [1/350][40/50] time 0.277 (0.295) data 0.000 (0.008) eta 1:25:48 loss 5.9363 (6.3801) acc 9.3750 (1.2500) lr 0.260000 -FPS@all 866.823, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.286 (0.294) data 0.000 (0.016) eta 1:25:42 loss 6.3082 (6.5779) acc 0.0000 (0.0000) lr 0.260000 -epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 6.0846 (6.4162) acc 3.1250 (0.6250) lr 0.260000 -FPS@all 866.665, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.284 (0.295) data 0.001 (0.012) eta 1:25:58 loss 6.7670 (6.5713) acc 3.1250 (0.4688) lr 0.260000 -epoch: [1/350][40/50] time 0.304 (0.296) data 0.000 (0.006) eta 1:26:06 loss 6.0424 (6.3826) acc 0.0000 (1.1719) lr 0.260000 -FPS@all 866.506, TIME@all 0.295 -group num: 1 -epoch: [1/350][20/50] time 0.285 (0.294) data 0.000 (0.016) eta 1:25:42 loss 6.5851 (6.6007) acc 0.0000 (0.9375) lr 0.260000 -epoch: [1/350][40/50] time 0.285 (0.295) data 0.000 (0.008) eta 1:25:48 loss 5.9063 (6.4240) acc 3.1250 (1.2500) lr 0.260000 -FPS@all 866.753, TIME@all 0.295 -group num: 1 -epoch: [1/1][20/50] time 0.236 (0.294) data 0.000 (0.015) eta 0:00:08 loss 6.5861 (6.5515) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/1][40/50] time 0.274 (0.294) data 0.001 (0.008) eta 0:00:02 loss 5.7942 (6.3841) acc 3.1250 (1.5625) lr 0.260000 -FPS@all 867.641, TIME@all 0.295 -Elapsed 0:07:45 -FPS@all 867.641, TIME@all 0.295 -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -terminate called after throwing an instance of 'std::runtime_error' - what(): AllReduce error in:/usr1/workspace/FPTA_Daily_open_2.0.3.tr5/CODE/pytorch/torch/lib/c10d/ProcessGroupHCCL.cpp: 114 -/root/archiconda3/envs/ych/lib/python3.7/multiprocessing/semaphore_tracker.py:144: UserWarning: semaphore_tracker: There appear to be 91 leaked semaphores to clean up at shutdown - len(cache)) -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Traceback (most recent call last): - File "main.py", line 55, in - import torch_npu -ModuleNotFoundError: No module named 'torch_npu' -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 6 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 0 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 2 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 1 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 3 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 4 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 5 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 7 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. - -Defaults for this optimization level are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : dynamic -combine_grad : None -combine_ddp : None -ddp_replica_count : 4 -check_combined_tensors : None -user_cast_preferred : None -Processing user overrides (additional kwargs that are not None)... -After processing overrides, optimization options are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : 128.0 -combine_grad : True -combine_ddp : None -ddp_replica_count : 4 -check_combined_tensors : None -user_cast_preferred : None -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.020) eta 1:28:17 loss 6.4503 (6.6220) acc 0.0000 (0.9375) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:18 loss 6.1957 (6.4520) acc 0.0000 (1.2500) lr 0.260000 -FPS@all 844.099, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.021) eta 1:28:17 loss 6.4964 (6.6583) acc 0.0000 (0.1562) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:28:18 loss 6.1912 (6.4751) acc 3.1250 (0.7031) lr 0.260000 -FPS@all 844.086, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.021) eta 1:28:17 loss 6.5021 (6.5814) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:28:18 loss 5.8930 (6.4415) acc 9.3750 (1.1719) lr 0.260000 -FPS@all 844.105, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.019) eta 1:28:16 loss 6.6479 (6.6206) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.001 (0.010) eta 1:28:18 loss 6.2197 (6.4612) acc 0.0000 (1.0938) lr 0.260000 -FPS@all 844.124, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.000 (0.020) eta 1:28:17 loss 6.7775 (6.5672) acc 0.0000 (0.6250) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:18 loss 6.0981 (6.4246) acc 3.1250 (1.1719) lr 0.260000 -FPS@all 844.117, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.001 (0.020) eta 1:28:15 loss 6.6308 (6.5626) acc 0.0000 (0.1562) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:17 loss 6.0067 (6.4354) acc 3.1250 (0.9375) lr 0.260000 -FPS@all 844.289, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.305 (0.303) data 0.001 (0.020) eta 1:28:15 loss 6.5050 (6.6098) acc 0.0000 (0.7812) lr 0.260000 -epoch: [1/350][40/50] time 0.303 (0.303) data 0.000 (0.010) eta 1:28:17 loss 6.1065 (6.4767) acc 3.1250 (0.7812) lr 0.260000 -FPS@all 844.264, TIME@all 0.303 -group num: 1 -epoch: [1/350][20/50] time 0.304 (0.303) data 0.000 (0.020) eta 1:28:16 loss 6.3586 (6.5710) acc 0.0000 (0.6250) lr 0.260000 -epoch: [1/350][40/50] time 0.304 (0.303) data 0.000 (0.010) eta 1:28:18 loss 5.9643 (6.4363) acc 6.2500 (0.8594) lr 0.260000 -FPS@all 844.485, TIME@all 0.303 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:43 loss 5.6904 (5.6815) acc 0.0000 (3.1250) lr 0.260000 -epoch: [2/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.6236 (5.6524) acc 6.2500 (3.3594) lr 0.260000 -FPS@all 841.331, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.9472 (5.6783) acc 3.1250 (4.3750) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:27 loss 6.0357 (5.6351) acc 3.1250 (4.9219) lr 0.260000 -FPS@all 841.359, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:42 loss 5.8032 (5.7169) acc 0.0000 (4.5312) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.3269 (5.5991) acc 9.3750 (5.5469) lr 0.260000 -FPS@all 841.341, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.9521 (5.6746) acc 3.1250 (4.0625) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.2284 (5.5886) acc 6.2500 (4.6875) lr 0.260000 -FPS@all 841.393, TIME@all 0.304 -epoch: [2/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:28:41 loss 5.7224 (5.6761) acc 6.2500 (3.2812) lr 0.260000 -epoch: [2/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:28:27 loss 5.2207 (5.6229) acc 3.1250 (4.4531) lr 0.260000 -FPS@all 841.667, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:41 loss 6.0267 (5.6724) acc 3.1250 (2.3438) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:28:26 loss 5.7390 (5.6345) acc 9.3750 (3.6719) lr 0.260000 -FPS@all 841.543, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:28:42 loss 5.7242 (5.7595) acc 3.1250 (2.5000) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:28:28 loss 5.2590 (5.6375) acc 3.1250 (3.5156) lr 0.260000 -FPS@all 841.338, TIME@all 0.304 -epoch: [2/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:28:42 loss 6.2207 (5.7487) acc 0.0000 (3.4375) lr 0.260000 -epoch: [2/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:28:27 loss 5.6020 (5.6702) acc 6.2500 (4.3750) lr 0.260000 -FPS@all 841.475, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:27:57 loss 5.1104 (4.9978) acc 9.3750 (9.5312) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:27:59 loss 4.9897 (4.9629) acc 9.3750 (9.6875) lr 0.260000 -FPS@all 842.763, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:27:58 loss 5.0472 (4.9345) acc 15.6250 (10.0000) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:28:00 loss 4.9986 (4.9421) acc 12.5000 (10.7812) lr 0.260000 -FPS@all 842.666, TIME@all 0.304 -epoch: [3/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:27:58 loss 5.1691 (4.9396) acc 9.3750 (9.0625) lr 0.260000 -epoch: [3/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:27:59 loss 4.5438 (4.9401) acc 15.6250 (10.7031) lr 0.260000 -FPS@all 842.701, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:27:57 loss 5.0295 (4.9700) acc 9.3750 (7.0312) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:27:58 loss 4.8667 (5.0102) acc 9.3750 (8.2031) lr 0.260000 -FPS@all 842.872, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:58 loss 5.0562 (4.9847) acc 3.1250 (8.5938) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:28:00 loss 4.4445 (4.9519) acc 25.0000 (11.0156) lr 0.260000 -FPS@all 842.679, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:58 loss 4.8639 (5.0062) acc 6.2500 (10.3125) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:28:00 loss 4.6740 (4.9698) acc 9.3750 (10.9375) lr 0.260000 -FPS@all 842.677, TIME@all 0.304 -epoch: [3/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:27:57 loss 5.1656 (4.9917) acc 9.3750 (9.0625) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:27:59 loss 4.8903 (4.9856) acc 6.2500 (9.6094) lr 0.260000 -FPS@all 842.807, TIME@all 0.304 -epoch: [3/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:58 loss 5.4447 (5.0193) acc 0.0000 (7.5000) lr 0.260000 -epoch: [3/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:28:00 loss 4.6572 (5.0107) acc 15.6250 (9.4531) lr 0.260000 -FPS@all 843.073, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:27:49 loss 4.5347 (4.3350) acc 25.0000 (17.8125) lr 0.260000 -epoch: [4/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:27:49 loss 4.2485 (4.3576) acc 15.6250 (18.9844) lr 0.260000 -FPS@all 841.958, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:48 loss 4.5189 (4.3321) acc 21.8750 (18.4375) lr 0.260000 -epoch: [4/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.7004 (4.3285) acc 15.6250 (19.8438) lr 0.260000 -FPS@all 842.039, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.3967 (4.3375) acc 15.6250 (18.1250) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 3.9165 (4.3720) acc 31.2500 (19.2188) lr 0.260000 -FPS@all 841.983, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.3063 (4.3083) acc 15.6250 (20.7812) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.2638 (4.3564) acc 18.7500 (20.5469) lr 0.260000 -FPS@all 842.026, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:27:49 loss 4.1575 (4.2676) acc 31.2500 (21.8750) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:49 loss 4.7793 (4.3274) acc 21.8750 (19.6875) lr 0.260000 -FPS@all 841.969, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:27:48 loss 4.5746 (4.2994) acc 21.8750 (19.2188) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.3279 (4.4049) acc 25.0000 (17.8125) lr 0.260000 -FPS@all 842.127, TIME@all 0.304 -epoch: [4/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:27:47 loss 4.1815 (4.3585) acc 15.6250 (18.7500) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.4623 (4.3688) acc 25.0000 (19.7656) lr 0.260000 -FPS@all 842.311, TIME@all 0.304 -epoch: [4/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:27:47 loss 4.3963 (4.2453) acc 18.7500 (19.6875) lr 0.260000 -epoch: [4/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:48 loss 4.4185 (4.3179) acc 18.7500 (19.3750) lr 0.260000 -FPS@all 842.184, TIME@all 0.304 -epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:27:36 loss 4.0796 (3.6327) acc 25.0000 (32.9688) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 4.4161 (3.7696) acc 12.5000 (31.0938) lr 0.260000 -FPS@all 842.687, TIME@all 0.304 -epoch: [5/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:27:35 loss 4.3063 (3.6727) acc 21.8750 (30.0000) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 3.9690 (3.7655) acc 37.5000 (31.0156) lr 0.260000 -FPS@all 842.738, TIME@all 0.304 -epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:27:36 loss 4.2187 (3.6356) acc 25.0000 (32.1875) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:27:30 loss 3.6517 (3.7157) acc 37.5000 (32.5000) lr 0.260000 -FPS@all 842.732, TIME@all 0.304 -epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:36 loss 4.3091 (3.7495) acc 28.1250 (30.4688) lr 0.260000 -epoch: [5/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:30 loss 3.9241 (3.8058) acc 31.2500 (31.4062) lr 0.260000 -FPS@all 842.668, TIME@all 0.304 -epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:27:34 loss 3.1360 (3.6162) acc 37.5000 (34.6875) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.6961 (3.7158) acc 37.5000 (33.6719) lr 0.260000 -FPS@all 842.898, TIME@all 0.304 -epoch: [5/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:27:35 loss 3.7299 (3.6657) acc 25.0000 (32.1875) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:30 loss 3.3893 (3.7434) acc 50.0000 (33.1250) lr 0.260000 -FPS@all 842.690, TIME@all 0.304 -epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:34 loss 3.4656 (3.6112) acc 40.6250 (36.4062) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.8184 (3.7159) acc 37.5000 (35.7031) lr 0.260000 -FPS@all 842.850, TIME@all 0.304 -epoch: [5/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:27:35 loss 3.8033 (3.6672) acc 37.5000 (34.6875) lr 0.260000 -epoch: [5/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:29 loss 3.7822 (3.7599) acc 21.8750 (32.2656) lr 0.260000 -FPS@all 843.088, TIME@all 0.304 -epoch: [6/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:27:17 loss 3.4712 (3.2225) acc 37.5000 (42.3438) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.2355 (3.3175) acc 46.8750 (42.0312) lr 0.260000 -FPS@all 844.135, TIME@all 0.303 -epoch: [6/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 1:27:17 loss 3.5757 (3.2662) acc 28.1250 (42.8125) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.8138 (3.3923) acc 34.3750 (39.9219) lr 0.260000 -FPS@all 844.166, TIME@all 0.303 -epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:16 loss 3.7874 (3.2680) acc 34.3750 (43.4375) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:06 loss 3.4127 (3.3299) acc 43.7500 (41.8750) lr 0.260000 -FPS@all 844.297, TIME@all 0.303 -epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:17 loss 3.5408 (3.2467) acc 40.6250 (43.5938) lr 0.260000 -epoch: [6/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:27:07 loss 3.4385 (3.3440) acc 37.5000 (41.6406) lr 0.260000 -FPS@all 844.129, TIME@all 0.303 -epoch: [6/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:27:17 loss 3.5238 (3.1541) acc 34.3750 (46.2500) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:07 loss 3.0333 (3.2738) acc 50.0000 (44.6094) lr 0.260000 -FPS@all 844.151, TIME@all 0.303 -epoch: [6/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:27:16 loss 3.7588 (3.2462) acc 31.2500 (44.8438) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:27:06 loss 3.6809 (3.3569) acc 28.1250 (42.5781) lr 0.260000 -FPS@all 844.330, TIME@all 0.303 -epoch: [6/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:27:16 loss 3.6366 (3.2437) acc 31.2500 (41.4062) lr 0.260000 -epoch: [6/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.4791 (3.3068) acc 28.1250 (40.6250) lr 0.260000 -FPS@all 844.182, TIME@all 0.303 -epoch: [6/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:27:18 loss 3.3052 (3.1660) acc 43.7500 (44.8438) lr 0.260000 -epoch: [6/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:27:07 loss 3.6164 (3.2659) acc 40.6250 (43.0469) lr 0.260000 -FPS@all 844.436, TIME@all 0.303 -epoch: [7/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:50 loss 2.7394 (2.7289) acc 46.8750 (54.8438) lr 0.260000 -epoch: [7/350][40/50] time 0.306 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.2769 (2.9206) acc 43.7500 (50.7812) lr 0.260000 -FPS@all 844.501, TIME@all 0.303 -epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.6852 (2.7088) acc 53.1250 (58.7500) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.1574 (2.8643) acc 43.7500 (54.5312) lr 0.260000 -FPS@all 844.461, TIME@all 0.303 -epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 3.0215 (2.7730) acc 43.7500 (56.2500) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.2028 (2.9054) acc 43.7500 (53.6719) lr 0.260000 -FPS@all 844.508, TIME@all 0.303 -epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 1:26:48 loss 2.7774 (2.6933) acc 56.2500 (59.3750) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:45 loss 2.9467 (2.8475) acc 56.2500 (56.7969) lr 0.260000 -FPS@all 844.674, TIME@all 0.303 -epoch: [7/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 1:26:51 loss 3.0027 (2.7353) acc 50.0000 (54.5312) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.1123 (2.8863) acc 59.3750 (51.9531) lr 0.260000 -FPS@all 844.811, TIME@all 0.303 -epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.9843 (2.7426) acc 59.3750 (56.5625) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.3751 (2.8929) acc 53.1250 (53.9844) lr 0.260000 -FPS@all 844.469, TIME@all 0.303 -epoch: [7/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:48 loss 2.5921 (2.7193) acc 68.7500 (57.3438) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:45 loss 3.2694 (2.8827) acc 53.1250 (54.0625) lr 0.260000 -FPS@all 844.641, TIME@all 0.303 -epoch: [7/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:26:49 loss 2.8563 (2.7895) acc 56.2500 (56.5625) lr 0.260000 -epoch: [7/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 1:26:46 loss 3.7664 (2.9228) acc 37.5000 (52.8125) lr 0.260000 -FPS@all 844.499, TIME@all 0.303 -epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:27:00 loss 2.9792 (2.4935) acc 53.1250 (66.0938) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.8324 (2.6809) acc 56.2500 (60.6250) lr 0.260000 -FPS@all 842.667, TIME@all 0.304 -epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:26:59 loss 3.1148 (2.5098) acc 53.1250 (63.7500) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:26:45 loss 2.8392 (2.6727) acc 56.2500 (60.6250) lr 0.260000 -FPS@all 842.702, TIME@all 0.304 -epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:27:00 loss 3.0805 (2.5866) acc 50.0000 (60.3125) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:26:45 loss 2.7478 (2.7204) acc 59.3750 (56.6406) lr 0.260000 -FPS@all 842.592, TIME@all 0.304 -epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:26:59 loss 3.0695 (2.5173) acc 56.2500 (64.8438) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.6999 (2.6806) acc 59.3750 (60.8594) lr 0.260000 -FPS@all 842.623, TIME@all 0.304 -epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:26:58 loss 2.9053 (2.5056) acc 62.5000 (65.7812) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.5641 (2.6186) acc 59.3750 (62.1875) lr 0.260000 -FPS@all 842.831, TIME@all 0.304 -epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:26:59 loss 3.0868 (2.5277) acc 59.3750 (62.9688) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:45 loss 2.5176 (2.6838) acc 59.3750 (58.2031) lr 0.260000 -FPS@all 842.627, TIME@all 0.304 -epoch: [8/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:26:57 loss 3.1584 (2.4755) acc 53.1250 (66.0938) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.3734 (2.6147) acc 68.7500 (61.3281) lr 0.260000 -FPS@all 843.071, TIME@all 0.304 -epoch: [8/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:26:58 loss 3.1165 (2.4870) acc 50.0000 (64.6875) lr 0.260000 -epoch: [8/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:26:44 loss 2.7944 (2.6475) acc 53.1250 (59.5312) lr 0.260000 -FPS@all 842.780, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:26 loss 2.9404 (2.3622) acc 53.1250 (70.6250) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:26:24 loss 2.8924 (2.4852) acc 53.1250 (65.4688) lr 0.260000 -FPS@all 843.200, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:26:26 loss 3.0240 (2.3398) acc 56.2500 (70.4688) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.6576 (2.4789) acc 59.3750 (67.4219) lr 0.260000 -FPS@all 843.262, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:25 loss 3.2055 (2.3575) acc 53.1250 (69.0625) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:23 loss 2.7023 (2.5233) acc 59.3750 (65.0000) lr 0.260000 -FPS@all 843.411, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:25 loss 2.8743 (2.3435) acc 53.1250 (68.4375) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.8318 (2.4589) acc 59.3750 (65.4688) lr 0.260000 -FPS@all 843.267, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:26 loss 3.1199 (2.4655) acc 43.7500 (63.9062) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.9502 (2.5648) acc 56.2500 (62.9688) lr 0.260000 -FPS@all 843.218, TIME@all 0.304 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:26:25 loss 3.3551 (2.3044) acc 43.7500 (70.3125) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:23 loss 3.2092 (2.4686) acc 56.2500 (66.5625) lr 0.260000 -FPS@all 843.418, TIME@all 0.304 -epoch: [9/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:26:26 loss 3.4728 (2.3746) acc 50.0000 (66.4062) lr 0.260000 -epoch: [9/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:26:23 loss 3.1980 (2.4826) acc 43.7500 (64.7656) lr 0.260000 -FPS@all 843.549, TIME@all 0.303 -epoch: [9/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:26:26 loss 2.8887 (2.2948) acc 53.1250 (69.6875) lr 0.260000 -epoch: [9/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:26:24 loss 2.7275 (2.4272) acc 53.1250 (66.4062) lr 0.260000 -FPS@all 843.220, TIME@all 0.304 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.6988 (2.0792) acc 53.1250 (75.3125) lr 0.260000 -epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.1807 (2.2598) acc 78.1250 (70.9375) lr 0.260000 -FPS@all 844.438, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.4798 (2.0968) acc 68.7500 (78.5938) lr 0.260000 -epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.4168 (2.2385) acc 59.3750 (73.0469) lr 0.260000 -FPS@all 844.346, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.3467 (2.0726) acc 78.1250 (77.1875) lr 0.260000 -epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:26:04 loss 2.4815 (2.2307) acc 65.6250 (72.4219) lr 0.260000 -FPS@all 844.407, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.3339 (2.0752) acc 78.1250 (76.8750) lr 0.260000 -epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:26:04 loss 2.0649 (2.2017) acc 78.1250 (73.9844) lr 0.260000 -FPS@all 844.389, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:06 loss 2.4495 (2.0035) acc 71.8750 (80.6250) lr 0.260000 -epoch: [10/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.0464 (2.1636) acc 75.0000 (75.3125) lr 0.260000 -FPS@all 844.372, TIME@all 0.303 -epoch: [10/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 1:26:06 loss 2.1854 (2.0741) acc 78.1250 (76.2500) lr 0.260000 -epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:04 loss 2.2622 (2.2498) acc 71.8750 (72.5781) lr 0.260000 -FPS@all 844.738, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:26:05 loss 2.3676 (2.1259) acc 71.8750 (74.3750) lr 0.260000 -epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:26:03 loss 2.4101 (2.2517) acc 75.0000 (72.8906) lr 0.260000 -FPS@all 844.590, TIME@all 0.303 -epoch: [10/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:26:05 loss 2.5710 (2.1373) acc 62.5000 (75.6250) lr 0.260000 -epoch: [10/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:26:03 loss 2.5480 (2.2394) acc 68.7500 (74.1406) lr 0.260000 -FPS@all 844.531, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 1:25:58 loss 2.3500 (1.9106) acc 71.8750 (81.7188) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 1.9245 (2.1120) acc 78.1250 (75.9375) lr 0.260000 -FPS@all 843.800, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.4370 (1.8808) acc 56.2500 (82.3438) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.3339 (2.0881) acc 71.8750 (77.3438) lr 0.260000 -FPS@all 843.859, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:58 loss 2.0455 (1.8739) acc 78.1250 (82.3438) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.4180 (2.0505) acc 68.7500 (78.4375) lr 0.260000 -FPS@all 843.821, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.1906 (1.9447) acc 75.0000 (80.0000) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.2909 (2.0804) acc 75.0000 (76.9531) lr 0.260000 -FPS@all 843.824, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:25:56 loss 2.1193 (1.9145) acc 75.0000 (84.3750) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:50 loss 2.0402 (2.1367) acc 78.1250 (76.9531) lr 0.260000 -FPS@all 844.009, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 1.9884 (1.8950) acc 75.0000 (81.5625) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:51 loss 2.4517 (2.1145) acc 65.6250 (76.6406) lr 0.260000 -FPS@all 843.813, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:25:57 loss 2.3042 (1.9266) acc 68.7500 (80.6250) lr 0.260000 -epoch: [11/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:50 loss 2.2633 (2.1020) acc 68.7500 (76.0156) lr 0.260000 -FPS@all 843.954, TIME@all 0.303 -epoch: [11/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:25:57 loss 2.5468 (1.9712) acc 56.2500 (80.4688) lr 0.260000 -epoch: [11/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:50 loss 2.0838 (2.0805) acc 78.1250 (77.4219) lr 0.260000 -FPS@all 844.160, TIME@all 0.303 -epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:50 loss 2.3477 (1.9139) acc 68.7500 (82.6562) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.1936 (2.0725) acc 68.7500 (77.9688) lr 0.260000 -FPS@all 842.655, TIME@all 0.304 -epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:50 loss 2.1892 (1.9188) acc 84.3750 (82.0312) lr 0.260000 -epoch: [12/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:25:46 loss 2.3637 (2.0791) acc 68.7500 (77.4219) lr 0.260000 -FPS@all 842.589, TIME@all 0.304 -epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:50 loss 2.1258 (1.9314) acc 84.3750 (81.2500) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.3672 (2.0763) acc 65.6250 (76.4062) lr 0.260000 -FPS@all 842.590, TIME@all 0.304 -epoch: [12/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.5196 (1.9502) acc 59.3750 (80.1562) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:45 loss 2.3533 (2.0755) acc 59.3750 (76.9531) lr 0.260000 -FPS@all 842.808, TIME@all 0.304 -epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:25:49 loss 2.0882 (1.9053) acc 78.1250 (83.5938) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:46 loss 2.2517 (2.0403) acc 71.8750 (79.2969) lr 0.260000 -FPS@all 842.615, TIME@all 0.304 -epoch: [12/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:25:50 loss 2.1714 (1.8672) acc 78.1250 (83.4375) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:25:46 loss 2.2614 (2.1188) acc 71.8750 (76.0938) lr 0.260000 -FPS@all 842.606, TIME@all 0.304 -epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.4305 (1.9507) acc 68.7500 (80.9375) lr 0.260000 -epoch: [12/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:25:45 loss 2.3816 (2.1251) acc 68.7500 (76.6406) lr 0.260000 -FPS@all 842.759, TIME@all 0.304 -epoch: [12/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:25:49 loss 2.0246 (1.8819) acc 78.1250 (84.2188) lr 0.260000 -epoch: [12/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:25:45 loss 1.9955 (2.0819) acc 75.0000 (78.8281) lr 0.260000 -FPS@all 842.996, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:28 loss 2.1344 (1.8721) acc 78.1250 (83.9062) lr 0.260000 -epoch: [13/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0766 (1.9707) acc 75.0000 (79.7656) lr 0.260000 -FPS@all 842.959, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:29 loss 2.2122 (1.8762) acc 62.5000 (82.5000) lr 0.260000 -epoch: [13/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0595 (1.9928) acc 75.0000 (80.4688) lr 0.260000 -FPS@all 842.901, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 1:25:29 loss 2.2509 (1.8982) acc 71.8750 (83.7500) lr 0.260000 -epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.2267 (1.9728) acc 75.0000 (80.7031) lr 0.260000 -FPS@all 842.874, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:29 loss 2.2561 (1.9078) acc 68.7500 (82.3438) lr 0.260000 -epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:25:26 loss 2.1083 (1.9869) acc 71.8750 (80.0781) lr 0.260000 -FPS@all 842.884, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:25:27 loss 2.0696 (1.8892) acc 75.0000 (82.9688) lr 0.260000 -epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:25:25 loss 1.9480 (1.9883) acc 84.3750 (79.1406) lr 0.260000 -FPS@all 843.092, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:29 loss 2.2525 (1.8713) acc 84.3750 (85.3125) lr 0.260000 -epoch: [13/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:25:26 loss 1.9996 (1.9672) acc 78.1250 (81.4062) lr 0.260000 -FPS@all 842.893, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:25:28 loss 2.0159 (1.8965) acc 81.2500 (84.2188) lr 0.260000 -epoch: [13/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:25:25 loss 1.9084 (1.9571) acc 84.3750 (82.0312) lr 0.260000 -FPS@all 843.018, TIME@all 0.304 -epoch: [13/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:25:27 loss 2.2537 (1.8796) acc 71.8750 (82.3438) lr 0.260000 -epoch: [13/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:25:26 loss 2.0106 (1.9679) acc 81.2500 (80.2344) lr 0.260000 -FPS@all 843.231, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.011) eta 1:25:39 loss 2.0438 (1.7822) acc 81.2500 (86.0938) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.0686 (1.8955) acc 75.0000 (82.4219) lr 0.260000 -FPS@all 842.934, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.001 (0.011) eta 1:25:38 loss 2.1837 (1.7730) acc 75.0000 (87.8125) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.1743 (1.8804) acc 78.1250 (84.0625) lr 0.260000 -FPS@all 842.978, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.011) eta 1:25:38 loss 2.2646 (1.7770) acc 71.8750 (86.2500) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.2471 (1.9031) acc 75.0000 (82.9688) lr 0.260000 -FPS@all 843.044, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3168 (1.7839) acc 75.0000 (84.3750) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.3512 (1.8856) acc 62.5000 (81.4062) lr 0.260000 -FPS@all 842.962, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3447 (1.8452) acc 71.8750 (82.6562) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.0083 (1.9111) acc 78.1250 (81.9531) lr 0.260000 -FPS@all 842.955, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:25:37 loss 2.2072 (1.7927) acc 71.8750 (87.3438) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:13 loss 1.8632 (1.8792) acc 81.2500 (83.6719) lr 0.260000 -FPS@all 843.169, TIME@all 0.304 -epoch: [14/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:25:38 loss 1.9521 (1.7738) acc 87.5000 (86.0938) lr 0.260000 -epoch: [14/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:25:14 loss 2.2463 (1.8876) acc 71.8750 (82.2656) lr 0.260000 -FPS@all 843.332, TIME@all 0.304 -epoch: [14/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:25:38 loss 2.3742 (1.7801) acc 68.7500 (85.9375) lr 0.260000 -epoch: [14/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:25:13 loss 1.9280 (1.8951) acc 75.0000 (82.0312) lr 0.260000 -FPS@all 843.106, TIME@all 0.304 -epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.011) eta 1:24:43 loss 1.9269 (1.7250) acc 90.6250 (89.2188) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 2.0384 (1.8059) acc 75.0000 (85.5469) lr 0.260000 -FPS@all 844.099, TIME@all 0.303 -epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.9626 (1.7266) acc 78.1250 (89.0625) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.6979 (1.8216) acc 93.7500 (85.2344) lr 0.260000 -FPS@all 844.128, TIME@all 0.303 -epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.011) eta 1:24:42 loss 1.6664 (1.7338) acc 87.5000 (87.0312) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 2.1876 (1.8102) acc 78.1250 (84.2969) lr 0.260000 -FPS@all 844.144, TIME@all 0.303 -epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.9600 (1.7508) acc 78.1250 (87.3438) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 1.8862 (1.8121) acc 81.2500 (85.0000) lr 0.260000 -FPS@all 844.110, TIME@all 0.303 -epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:24:42 loss 1.6719 (1.7434) acc 90.6250 (87.8125) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:24:46 loss 1.7332 (1.8116) acc 93.7500 (85.2344) lr 0.260000 -FPS@all 844.321, TIME@all 0.303 -epoch: [15/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:24:42 loss 1.8860 (1.7333) acc 84.3750 (87.6562) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.5067 (1.8059) acc 96.8750 (85.5469) lr 0.260000 -FPS@all 844.269, TIME@all 0.303 -epoch: [15/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 1:24:43 loss 1.7712 (1.7148) acc 87.5000 (87.5000) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:48 loss 1.6564 (1.7955) acc 93.7500 (84.8438) lr 0.260000 -FPS@all 844.099, TIME@all 0.303 -epoch: [15/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 1:24:42 loss 1.7334 (1.7192) acc 87.5000 (89.6875) lr 0.260000 -epoch: [15/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:47 loss 1.7172 (1.8116) acc 87.5000 (85.4688) lr 0.260000 -FPS@all 844.484, TIME@all 0.303 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.8232 (1.6718) acc 84.3750 (89.6875) lr 0.260000 -epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.7307 (1.7648) acc 87.5000 (85.7031) lr 0.260000 -FPS@all 842.654, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 2.2704 (1.7408) acc 68.7500 (85.9375) lr 0.260000 -epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.8284 (1.8023) acc 84.3750 (84.2188) lr 0.260000 -FPS@all 842.709, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.9320 (1.6923) acc 87.5000 (89.3750) lr 0.260000 -epoch: [16/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:24:43 loss 2.0041 (1.7521) acc 78.1250 (87.5781) lr 0.260000 -FPS@all 842.681, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:53 loss 1.9430 (1.6440) acc 90.6250 (91.7188) lr 0.260000 -epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:24:43 loss 1.8862 (1.7211) acc 78.1250 (88.5156) lr 0.260000 -FPS@all 842.678, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:52 loss 2.0422 (1.7293) acc 81.2500 (88.7500) lr 0.260000 -epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:42 loss 1.8093 (1.7598) acc 87.5000 (87.1875) lr 0.260000 -FPS@all 842.837, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:52 loss 1.9518 (1.6579) acc 78.1250 (88.5938) lr 0.260000 -epoch: [16/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:42 loss 1.8418 (1.7446) acc 84.3750 (86.0156) lr 0.260000 -FPS@all 842.879, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:53 loss 1.7201 (1.6666) acc 84.3750 (87.0312) lr 0.260000 -epoch: [16/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:24:43 loss 1.7872 (1.7213) acc 84.3750 (86.1719) lr 0.260000 -FPS@all 842.710, TIME@all 0.304 -epoch: [16/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:24:52 loss 1.6927 (1.6364) acc 87.5000 (88.2812) lr 0.260000 -epoch: [16/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:24:42 loss 1.9825 (1.7273) acc 68.7500 (85.7812) lr 0.260000 -FPS@all 843.087, TIME@all 0.304 -epoch: [17/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6068 (1.6308) acc 90.6250 (91.0938) lr 0.260000 -epoch: [17/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.9334 (1.6947) acc 75.0000 (88.3594) lr 0.260000 -FPS@all 842.256, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.8528 (1.6177) acc 87.5000 (90.0000) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:35 loss 2.1174 (1.7142) acc 71.8750 (87.3438) lr 0.260000 -FPS@all 842.178, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:37 loss 1.7029 (1.6356) acc 87.5000 (90.6250) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:24:33 loss 2.0299 (1.7207) acc 84.3750 (87.9688) lr 0.260000 -FPS@all 842.390, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:38 loss 1.7138 (1.6227) acc 84.3750 (91.0938) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.9223 (1.7092) acc 75.0000 (87.5000) lr 0.260000 -FPS@all 842.187, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6418 (1.6722) acc 90.6250 (89.2188) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.8367 (1.7258) acc 84.3750 (87.5000) lr 0.260000 -FPS@all 842.534, TIME@all 0.304 -epoch: [17/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6256 (1.6194) acc 87.5000 (90.9375) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.6249 (1.7132) acc 84.3750 (88.3594) lr 0.260000 -FPS@all 842.212, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:24:37 loss 1.8583 (1.6009) acc 81.2500 (91.0938) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:24:34 loss 1.9272 (1.6970) acc 87.5000 (88.2812) lr 0.260000 -FPS@all 842.352, TIME@all 0.304 -epoch: [17/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:24:38 loss 1.6337 (1.6059) acc 87.5000 (90.0000) lr 0.260000 -epoch: [17/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:24:34 loss 1.7891 (1.6995) acc 81.2500 (87.5781) lr 0.260000 -FPS@all 842.202, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:16 loss 1.6517 (1.6342) acc 87.5000 (89.8438) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:24:13 loss 1.5770 (1.6491) acc 84.3750 (89.2188) lr 0.260000 -FPS@all 841.871, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.8928 (1.6250) acc 84.3750 (90.3125) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:13 loss 1.6660 (1.6828) acc 93.7500 (88.5938) lr 0.260000 -FPS@all 841.909, TIME@all 0.304 -epoch: [18/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.8458 (1.6233) acc 84.3750 (90.1562) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.5797 (1.6621) acc 90.6250 (88.3594) lr 0.260000 -FPS@all 841.924, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.6356 (1.5906) acc 93.7500 (92.5000) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.4996 (1.6321) acc 96.8750 (90.3906) lr 0.260000 -FPS@all 841.863, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:16 loss 1.7957 (1.6170) acc 84.3750 (89.0625) lr 0.260000 -epoch: [18/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:24:13 loss 1.5924 (1.6327) acc 90.6250 (89.6875) lr 0.260000 -FPS@all 841.871, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:24:15 loss 1.6906 (1.6183) acc 90.6250 (89.3750) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:24:12 loss 1.5967 (1.6827) acc 87.5000 (87.2656) lr 0.260000 -FPS@all 842.051, TIME@all 0.304 -epoch: [18/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:15 loss 1.8373 (1.6506) acc 87.5000 (90.1562) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:12 loss 1.6518 (1.6625) acc 87.5000 (90.0000) lr 0.260000 -FPS@all 841.997, TIME@all 0.304 -epoch: [18/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:24:16 loss 2.0002 (1.6281) acc 84.3750 (90.1562) lr 0.260000 -epoch: [18/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:24:13 loss 1.6149 (1.6387) acc 90.6250 (89.9219) lr 0.260000 -FPS@all 842.202, TIME@all 0.304 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:59 loss 1.6540 (1.5709) acc 90.6250 (90.7812) lr 0.260000 -epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:23:51 loss 1.6243 (1.6345) acc 84.3750 (88.4375) lr 0.260000 -FPS@all 844.138, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:00 loss 1.6576 (1.5697) acc 84.3750 (90.3125) lr 0.260000 -epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.6060 (1.6423) acc 90.6250 (89.0625) lr 0.260000 -FPS@all 844.159, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:24:00 loss 1.8525 (1.5766) acc 81.2500 (90.3125) lr 0.260000 -epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:23:51 loss 1.4212 (1.6297) acc 96.8750 (89.0625) lr 0.260000 -FPS@all 844.170, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:59 loss 2.1074 (1.6043) acc 75.0000 (90.6250) lr 0.260000 -epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.5387 (1.6736) acc 90.6250 (88.2031) lr 0.260000 -FPS@all 844.314, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:23:58 loss 1.8209 (1.5300) acc 78.1250 (92.9688) lr 0.260000 -epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.4963 (1.5918) acc 93.7500 (91.8750) lr 0.260000 -FPS@all 844.358, TIME@all 0.303 -epoch: [19/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:23:57 loss 1.8268 (1.5641) acc 84.3750 (92.1875) lr 0.260000 -epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:50 loss 1.6087 (1.6355) acc 87.5000 (89.0625) lr 0.260000 -FPS@all 844.445, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:24:00 loss 1.6196 (1.5919) acc 87.5000 (90.7812) lr 0.260000 -epoch: [19/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.5862 (1.6471) acc 90.6250 (88.9844) lr 0.260000 -FPS@all 844.171, TIME@all 0.303 -epoch: [19/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:59 loss 1.7431 (1.5879) acc 78.1250 (89.0625) lr 0.260000 -epoch: [19/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:23:51 loss 1.8701 (1.6353) acc 81.2500 (88.7500) lr 0.260000 -FPS@all 844.173, TIME@all 0.303 -epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6368 (1.5870) acc 90.6250 (90.6250) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.4452 (1.5943) acc 100.0000 (90.5469) lr 0.260000 -FPS@all 843.509, TIME@all 0.303 -epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:49 loss 1.6363 (1.5571) acc 90.6250 (92.3438) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.6918 (1.6112) acc 90.6250 (90.0781) lr 0.260000 -FPS@all 843.468, TIME@all 0.304 -epoch: [20/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 1:23:49 loss 1.5494 (1.5056) acc 93.7500 (92.8125) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.7375 (1.5875) acc 84.3750 (90.1562) lr 0.260000 -FPS@all 843.395, TIME@all 0.304 -epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:49 loss 1.6179 (1.5562) acc 90.6250 (91.8750) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.4542 (1.6301) acc 96.8750 (89.7656) lr 0.260000 -FPS@all 843.438, TIME@all 0.304 -epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:23:48 loss 1.7041 (1.5407) acc 87.5000 (90.7812) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:36 loss 1.7553 (1.5971) acc 87.5000 (89.6875) lr 0.260000 -FPS@all 843.632, TIME@all 0.303 -epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6396 (1.5228) acc 93.7500 (93.1250) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.5644 (1.5941) acc 90.6250 (91.0938) lr 0.260000 -FPS@all 843.447, TIME@all 0.304 -epoch: [20/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.6168 (1.5395) acc 87.5000 (92.3438) lr 0.260000 -epoch: [20/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:23:37 loss 1.5754 (1.5801) acc 96.8750 (90.9375) lr 0.260000 -FPS@all 843.756, TIME@all 0.303 -epoch: [20/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:48 loss 1.7928 (1.5319) acc 84.3750 (90.9375) lr 0.260000 -epoch: [20/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:36 loss 1.6874 (1.6130) acc 90.6250 (89.2188) lr 0.260000 -FPS@all 843.565, TIME@all 0.303 -epoch: [21/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:23:32 loss 1.6609 (1.4867) acc 87.5000 (94.3750) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.5925 (1.5507) acc 90.6250 (92.0312) lr 0.260000 -FPS@all 842.783, TIME@all 0.304 -epoch: [21/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:33 loss 1.6264 (1.5211) acc 90.6250 (92.6562) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.7088 (1.5811) acc 87.5000 (91.0156) lr 0.260000 -FPS@all 842.739, TIME@all 0.304 -epoch: [21/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:23:32 loss 1.5246 (1.5038) acc 93.7500 (92.8125) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:23:23 loss 1.6952 (1.5606) acc 96.8750 (91.8750) lr 0.260000 -FPS@all 842.785, TIME@all 0.304 -epoch: [21/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.4524 (1.4895) acc 96.8750 (93.4375) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:24 loss 1.6472 (1.5544) acc 90.6250 (91.3281) lr 0.260000 -FPS@all 842.775, TIME@all 0.304 -epoch: [21/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:23:31 loss 1.6183 (1.5179) acc 87.5000 (92.5000) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:22 loss 1.7229 (1.5657) acc 81.2500 (91.5625) lr 0.260000 -FPS@all 842.969, TIME@all 0.304 -epoch: [21/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.3999 (1.5017) acc 96.8750 (92.9688) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:24 loss 1.6378 (1.5621) acc 87.5000 (91.5625) lr 0.260000 -FPS@all 842.771, TIME@all 0.304 -epoch: [21/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:23:32 loss 1.6774 (1.5058) acc 87.5000 (93.2812) lr 0.260000 -epoch: [21/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:23 loss 1.8060 (1.5426) acc 84.3750 (92.4219) lr 0.260000 -FPS@all 842.925, TIME@all 0.304 -epoch: [21/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:23:33 loss 1.5637 (1.5081) acc 90.6250 (93.1250) lr 0.260000 -epoch: [21/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:23:24 loss 1.6745 (1.5669) acc 84.3750 (91.6406) lr 0.260000 -FPS@all 843.104, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4547 (1.4388) acc 90.6250 (94.3750) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5501 (1.5267) acc 90.6250 (92.5781) lr 0.260000 -FPS@all 841.927, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4106 (1.4491) acc 96.8750 (94.5312) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5252 (1.5035) acc 96.8750 (93.2812) lr 0.260000 -FPS@all 841.989, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.4990 (1.4538) acc 93.7500 (94.8438) lr 0.260000 -epoch: [22/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:23:15 loss 1.4044 (1.5409) acc 96.8750 (92.5000) lr 0.260000 -FPS@all 841.962, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.5454 (1.4536) acc 90.6250 (94.3750) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:23:16 loss 1.5478 (1.5200) acc 93.7500 (92.9688) lr 0.260000 -FPS@all 841.960, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:23:24 loss 1.5106 (1.4572) acc 90.6250 (94.2188) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:14 loss 1.5968 (1.5300) acc 90.6250 (92.5000) lr 0.260000 -FPS@all 842.155, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:23:24 loss 1.6081 (1.5177) acc 93.7500 (93.2812) lr 0.260000 -epoch: [22/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:23:15 loss 1.7207 (1.5342) acc 84.3750 (93.0469) lr 0.260000 -FPS@all 842.318, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:23:25 loss 1.6046 (1.4892) acc 90.6250 (93.5938) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:16 loss 1.7763 (1.5437) acc 84.3750 (91.4844) lr 0.260000 -FPS@all 841.945, TIME@all 0.304 -epoch: [22/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:23:24 loss 1.4799 (1.4689) acc 90.6250 (95.3125) lr 0.260000 -epoch: [22/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:23:15 loss 1.5760 (1.5538) acc 90.6250 (91.8750) lr 0.260000 -FPS@all 842.080, TIME@all 0.304 -epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 1:22:49 loss 1.6826 (1.4862) acc 93.7500 (92.9688) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:47 loss 1.5364 (1.5273) acc 87.5000 (92.1094) lr 0.260000 -FPS@all 844.241, TIME@all 0.303 -epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 1:22:48 loss 1.3989 (1.4764) acc 96.8750 (92.9688) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:46 loss 1.4861 (1.5280) acc 96.8750 (91.7969) lr 0.260000 -FPS@all 844.320, TIME@all 0.303 -epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:22:49 loss 1.5566 (1.4916) acc 90.6250 (92.9688) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5078 (1.5196) acc 93.7500 (92.0312) lr 0.260000 -FPS@all 844.275, TIME@all 0.303 -epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.5836 (1.4664) acc 93.7500 (94.2188) lr 0.260000 -epoch: [23/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5335 (1.5126) acc 90.6250 (92.6562) lr 0.260000 -FPS@all 844.281, TIME@all 0.303 -epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 1:22:47 loss 1.5609 (1.4499) acc 90.6250 (95.3125) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:45 loss 1.5147 (1.5287) acc 90.6250 (92.7344) lr 0.260000 -FPS@all 844.469, TIME@all 0.303 -epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.4855 (1.5202) acc 90.6250 (92.5000) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.3572 (1.5464) acc 100.0000 (92.1875) lr 0.260000 -FPS@all 844.280, TIME@all 0.303 -epoch: [23/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.4994 (1.4266) acc 90.6250 (95.1562) lr 0.260000 -epoch: [23/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.6486 (1.5329) acc 87.5000 (92.1875) lr 0.260000 -FPS@all 844.569, TIME@all 0.303 -epoch: [23/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:22:48 loss 1.5796 (1.4849) acc 93.7500 (94.3750) lr 0.260000 -epoch: [23/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:46 loss 1.5325 (1.5279) acc 87.5000 (92.2656) lr 0.260000 -FPS@all 844.394, TIME@all 0.303 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 1:22:47 loss 1.4438 (1.4761) acc 93.7500 (93.2812) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:42 loss 1.6138 (1.5045) acc 90.6250 (92.1094) lr 0.260000 -FPS@all 843.258, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.5758 (1.4738) acc 90.6250 (93.5938) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.4847 (1.5046) acc 93.7500 (92.7344) lr 0.260000 -FPS@all 843.292, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 1:22:46 loss 1.4917 (1.5130) acc 93.7500 (92.8125) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:42 loss 1.5912 (1.5180) acc 93.7500 (92.8125) lr 0.260000 -FPS@all 843.313, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.4771 (1.4776) acc 90.6250 (93.4375) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.7465 (1.5324) acc 81.2500 (91.7969) lr 0.260000 -FPS@all 843.288, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 1:22:46 loss 1.5804 (1.4843) acc 90.6250 (93.2812) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.4862 (1.4840) acc 93.7500 (93.3594) lr 0.260000 -FPS@all 843.424, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:47 loss 1.6834 (1.4677) acc 93.7500 (94.3750) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:42 loss 1.4685 (1.5081) acc 93.7500 (93.0469) lr 0.260000 -FPS@all 843.268, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.001 (0.014) eta 1:22:45 loss 1.4580 (1.4679) acc 93.7500 (94.0625) lr 0.260000 -epoch: [24/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.7204 (1.5035) acc 84.3750 (92.8125) lr 0.260000 -FPS@all 843.483, TIME@all 0.304 -epoch: [24/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:22:46 loss 1.5345 (1.4554) acc 87.5000 (94.3750) lr 0.260000 -epoch: [24/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:22:41 loss 1.3837 (1.4919) acc 93.7500 (93.2031) lr 0.260000 -FPS@all 843.668, TIME@all 0.303 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5613 (1.4574) acc 93.7500 (95.1562) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.6645 (1.5084) acc 84.3750 (93.0469) lr 0.260000 -FPS@all 840.948, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:22:42 loss 1.6904 (1.5296) acc 90.6250 (92.5000) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:22:38 loss 1.5665 (1.5438) acc 93.7500 (91.4844) lr 0.260000 -FPS@all 840.910, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:22:41 loss 1.5668 (1.4762) acc 96.8750 (94.0625) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:22:37 loss 1.5616 (1.5334) acc 90.6250 (92.5000) lr 0.260000 -FPS@all 840.974, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5881 (1.4786) acc 87.5000 (93.5938) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:38 loss 1.4845 (1.4953) acc 87.5000 (92.7344) lr 0.260000 -FPS@all 840.924, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5575 (1.4775) acc 90.6250 (93.4375) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.5321 (1.5220) acc 90.6250 (92.4219) lr 0.260000 -FPS@all 841.077, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:41 loss 1.5106 (1.5379) acc 100.0000 (92.1875) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:38 loss 1.5961 (1.5574) acc 87.5000 (91.2500) lr 0.260000 -FPS@all 840.921, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:42 loss 1.5237 (1.4779) acc 93.7500 (94.8438) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:37 loss 1.5036 (1.5181) acc 93.7500 (93.4375) lr 0.260000 -FPS@all 841.256, TIME@all 0.304 -epoch: [25/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:22:40 loss 1.5909 (1.5067) acc 87.5000 (92.8125) lr 0.260000 -epoch: [25/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 1:22:36 loss 1.6199 (1.5342) acc 90.6250 (91.9531) lr 0.260000 -FPS@all 841.136, TIME@all 0.304 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:22:15 loss 1.4695 (1.4284) acc 90.6250 (95.1562) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:02 loss 1.5281 (1.4940) acc 93.7500 (93.2031) lr 0.260000 -FPS@all 844.639, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4233 (1.4277) acc 100.0000 (95.3125) lr 0.260000 -epoch: [26/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.6222 (1.4897) acc 87.5000 (93.5156) lr 0.260000 -FPS@all 844.690, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:22:14 loss 1.4912 (1.4271) acc 90.6250 (95.0000) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:22:02 loss 1.4413 (1.4866) acc 96.8750 (92.6562) lr 0.260000 -FPS@all 844.685, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4756 (1.4871) acc 90.6250 (93.2812) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.5794 (1.5400) acc 87.5000 (91.2500) lr 0.260000 -FPS@all 844.648, TIME@all 0.303 -epoch: [26/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4982 (1.4342) acc 93.7500 (95.3125) lr 0.260000 -epoch: [26/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:22:00 loss 1.3261 (1.4772) acc 96.8750 (93.7500) lr 0.260000 -FPS@all 845.017, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.4850 (1.4275) acc 96.8750 (94.8438) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:01 loss 1.4930 (1.4946) acc 93.7500 (92.6562) lr 0.260000 -FPS@all 844.810, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:14 loss 1.6054 (1.4459) acc 90.6250 (94.3750) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:01 loss 1.6510 (1.5111) acc 90.6250 (92.5781) lr 0.260000 -FPS@all 844.856, TIME@all 0.303 -epoch: [26/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:22:15 loss 1.3985 (1.4197) acc 90.6250 (93.2812) lr 0.260000 -epoch: [26/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:22:02 loss 1.4797 (1.4796) acc 96.8750 (92.3438) lr 0.260000 -FPS@all 844.669, TIME@all 0.303 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.5097 (1.4394) acc 90.6250 (94.5312) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:22:04 loss 1.4779 (1.4942) acc 93.7500 (92.9688) lr 0.260000 -FPS@all 842.440, TIME@all 0.304 -epoch: [27/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.4414 (1.4462) acc 96.8750 (93.9062) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.4625 (1.5148) acc 93.7500 (91.1719) lr 0.260000 -FPS@all 842.438, TIME@all 0.304 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:22:12 loss 1.5433 (1.4570) acc 90.6250 (94.5312) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:22:03 loss 1.5172 (1.4869) acc 90.6250 (93.8281) lr 0.260000 -FPS@all 842.525, TIME@all 0.304 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.4825 (1.4180) acc 93.7500 (95.6250) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.4554 (1.4910) acc 90.6250 (93.1250) lr 0.260000 -FPS@all 842.440, TIME@all 0.304 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:12 loss 1.5227 (1.4483) acc 93.7500 (93.2812) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.5613 (1.4768) acc 90.6250 (93.4375) lr 0.260000 -FPS@all 842.450, TIME@all 0.304 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:22:11 loss 1.4760 (1.4635) acc 90.6250 (93.5938) lr 0.260000 -epoch: [27/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:22:04 loss 1.7285 (1.5040) acc 84.3750 (92.1094) lr 0.260000 -FPS@all 842.816, TIME@all 0.304 -epoch: [27/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:22:11 loss 1.4941 (1.4683) acc 96.8750 (95.0000) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:03 loss 1.6245 (1.5023) acc 90.6250 (93.1250) lr 0.260000 -FPS@all 842.654, TIME@all 0.304 -epoch: [27/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:22:11 loss 1.6080 (1.4627) acc 87.5000 (94.5312) lr 0.260000 -epoch: [27/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:22:03 loss 1.4639 (1.5207) acc 96.8750 (92.1875) lr 0.260000 -FPS@all 842.614, TIME@all 0.304 -epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.5383 (1.4260) acc 87.5000 (94.6875) lr 0.260000 -epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.4964 (1.4809) acc 90.6250 (93.6719) lr 0.260000 -FPS@all 842.596, TIME@all 0.304 -epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:21:44 loss 1.3346 (1.4317) acc 96.8750 (95.1562) lr 0.260000 -epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3215 (1.4743) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 842.497, TIME@all 0.304 -epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:21:43 loss 1.3860 (1.4337) acc 96.8750 (94.5312) lr 0.260000 -epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.4900 (1.4702) acc 90.6250 (94.1406) lr 0.260000 -FPS@all 842.566, TIME@all 0.304 -epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:21:44 loss 1.3578 (1.4242) acc 96.8750 (94.6875) lr 0.260000 -epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3231 (1.4488) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 842.525, TIME@all 0.304 -epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.3018 (1.4068) acc 96.8750 (95.4688) lr 0.260000 -epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3695 (1.4556) acc 93.7500 (93.5156) lr 0.260000 -FPS@all 842.510, TIME@all 0.304 -epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:21:43 loss 1.2834 (1.4353) acc 96.8750 (94.5312) lr 0.260000 -epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:21:41 loss 1.4604 (1.4591) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 842.648, TIME@all 0.304 -epoch: [28/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:21:43 loss 1.3176 (1.4344) acc 100.0000 (95.1562) lr 0.260000 -epoch: [28/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:21:41 loss 1.3464 (1.4445) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 842.705, TIME@all 0.304 -epoch: [28/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:21:44 loss 1.4319 (1.4594) acc 96.8750 (94.8438) lr 0.260000 -epoch: [28/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:21:42 loss 1.3979 (1.4787) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 842.826, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 1:21:46 loss 1.4040 (1.4694) acc 93.7500 (94.6875) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.4826 (1.4985) acc 90.6250 (93.4375) lr 0.260000 -FPS@all 841.308, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:46 loss 1.5328 (1.4742) acc 93.7500 (93.9062) lr 0.260000 -epoch: [29/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.4068 (1.4879) acc 93.7500 (93.5938) lr 0.260000 -FPS@all 841.322, TIME@all 0.304 -epoch: [29/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:21:45 loss 1.3712 (1.4458) acc 96.8750 (94.6875) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:21:36 loss 1.5679 (1.5027) acc 93.7500 (92.5781) lr 0.260000 -FPS@all 841.347, TIME@all 0.304 -epoch: [29/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.2658 (1.4347) acc 100.0000 (95.1562) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:36 loss 1.4188 (1.4724) acc 96.8750 (93.5156) lr 0.260000 -FPS@all 841.316, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.4594 (1.4482) acc 93.7500 (94.2188) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:36 loss 1.5041 (1.4841) acc 93.7500 (93.4375) lr 0.260000 -FPS@all 841.332, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:44 loss 1.5366 (1.4618) acc 93.7500 (94.2188) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.3773 (1.4778) acc 93.7500 (93.5156) lr 0.260000 -FPS@all 841.709, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:21:45 loss 1.6286 (1.4389) acc 90.6250 (94.8438) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.3795 (1.4809) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 841.460, TIME@all 0.304 -epoch: [29/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 1:21:44 loss 1.4948 (1.4753) acc 87.5000 (93.9062) lr 0.260000 -epoch: [29/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 1:21:35 loss 1.4476 (1.4994) acc 96.8750 (92.8125) lr 0.260000 -FPS@all 841.514, TIME@all 0.304 -epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 1:21:13 loss 1.4448 (1.4722) acc 96.8750 (93.2812) lr 0.260000 -epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:21:10 loss 1.4200 (1.5036) acc 96.8750 (92.1875) lr 0.260000 -FPS@all 843.450, TIME@all 0.304 -epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 1:21:11 loss 1.5068 (1.4800) acc 93.7500 (93.9062) lr 0.260000 -epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:21:09 loss 1.4946 (1.5116) acc 93.7500 (92.7344) lr 0.260000 -FPS@all 843.592, TIME@all 0.303 -epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.6495 (1.4886) acc 90.6250 (94.8438) lr 0.260000 -epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:10 loss 1.4396 (1.5168) acc 93.7500 (93.2812) lr 0.260000 -FPS@all 843.478, TIME@all 0.304 -epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.7469 (1.4573) acc 90.6250 (94.6875) lr 0.260000 -epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 1:21:10 loss 1.4834 (1.4990) acc 93.7500 (93.3594) lr 0.260000 -FPS@all 843.474, TIME@all 0.304 -epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:12 loss 1.4647 (1.4640) acc 90.6250 (93.7500) lr 0.260000 -epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.4799 (1.5009) acc 100.0000 (92.5000) lr 0.260000 -FPS@all 843.658, TIME@all 0.303 -epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.6040 (1.4694) acc 87.5000 (93.7500) lr 0.260000 -epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.5660 (1.4948) acc 90.6250 (92.8125) lr 0.260000 -FPS@all 843.614, TIME@all 0.303 -epoch: [30/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:21:12 loss 1.4890 (1.4655) acc 90.6250 (94.2188) lr 0.260000 -epoch: [30/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:21:09 loss 1.3320 (1.5026) acc 100.0000 (92.6562) lr 0.260000 -FPS@all 843.809, TIME@all 0.303 -epoch: [30/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 1:21:13 loss 1.5546 (1.4493) acc 90.6250 (93.4375) lr 0.260000 -epoch: [30/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:21:09 loss 1.4316 (1.5283) acc 96.8750 (91.8750) lr 0.260000 -FPS@all 843.511, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:21:04 loss 1.4819 (1.3867) acc 96.8750 (97.0312) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:20:54 loss 1.4693 (1.4344) acc 90.6250 (94.7656) lr 0.260000 -FPS@all 843.471, TIME@all 0.304 -epoch: [31/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.5075 (1.4360) acc 93.7500 (94.6875) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:20:54 loss 1.3717 (1.4532) acc 100.0000 (94.1406) lr 0.260000 -FPS@all 843.540, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:21:04 loss 1.6758 (1.4415) acc 90.6250 (95.4688) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:20:54 loss 1.2680 (1.4672) acc 100.0000 (94.0625) lr 0.260000 -FPS@all 843.515, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.4401 (1.4132) acc 87.5000 (94.2188) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.4437 (1.4692) acc 96.8750 (93.2031) lr 0.260000 -FPS@all 843.633, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:05 loss 1.4271 (1.4243) acc 93.7500 (93.4375) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:20:55 loss 1.4306 (1.4675) acc 96.8750 (92.8906) lr 0.260000 -FPS@all 843.463, TIME@all 0.304 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:03 loss 1.5094 (1.4319) acc 93.7500 (94.8438) lr 0.260000 -epoch: [31/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.4957 (1.4566) acc 93.7500 (94.0625) lr 0.260000 -FPS@all 843.669, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 1:21:05 loss 1.4201 (1.4312) acc 93.7500 (95.3125) lr 0.260000 -epoch: [31/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:20:53 loss 1.3450 (1.4602) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 843.810, TIME@all 0.303 -epoch: [31/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:21:04 loss 1.4074 (1.4278) acc 96.8750 (94.8438) lr 0.260000 -epoch: [31/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:20:54 loss 1.4347 (1.4551) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 843.464, TIME@all 0.304 -epoch: [32/350][20/50] time 0.308 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.3327 (1.3897) acc 100.0000 (95.7812) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.6014 (1.4235) acc 87.5000 (94.5312) lr 0.260000 -FPS@all 843.763, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.011) eta 1:20:33 loss 1.5115 (1.4345) acc 90.6250 (93.9062) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.3753 (1.4714) acc 100.0000 (92.9688) lr 0.260000 -FPS@all 843.790, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.011) eta 1:20:33 loss 1.4833 (1.3992) acc 96.8750 (95.3125) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.4745 (1.4309) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 843.718, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.001 (0.012) eta 1:20:33 loss 1.8234 (1.4170) acc 90.6250 (95.7812) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:32 loss 1.3378 (1.4278) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 843.788, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.6402 (1.4190) acc 90.6250 (94.2188) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.3956 (1.4378) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 843.775, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.013) eta 1:20:32 loss 1.7985 (1.4069) acc 84.3750 (95.4688) lr 0.260000 -epoch: [32/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:31 loss 1.3286 (1.4361) acc 100.0000 (95.0000) lr 0.260000 -FPS@all 843.975, TIME@all 0.303 -epoch: [32/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 1:20:34 loss 1.4344 (1.3838) acc 93.7500 (96.5625) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:33 loss 1.5029 (1.4188) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 844.054, TIME@all 0.303 -epoch: [32/350][20/50] time 0.307 (0.303) data 0.000 (0.012) eta 1:20:33 loss 1.5059 (1.3754) acc 93.7500 (96.4062) lr 0.260000 -epoch: [32/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:20:32 loss 1.3600 (1.4100) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 843.893, TIME@all 0.303 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.2916 (1.3813) acc 96.8750 (96.0938) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:23 loss 1.5375 (1.4250) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 842.746, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.2976 (1.4035) acc 100.0000 (96.0938) lr 0.260000 -epoch: [33/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.2956 (1.4352) acc 100.0000 (94.9219) lr 0.260000 -FPS@all 842.779, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:20:32 loss 1.3914 (1.3628) acc 93.7500 (96.7188) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.4482 (1.4207) acc 96.8750 (95.3906) lr 0.260000 -FPS@all 842.720, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:20:32 loss 1.5378 (1.3885) acc 90.6250 (95.6250) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:20:23 loss 1.3678 (1.4402) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 842.734, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:31 loss 1.3564 (1.3609) acc 96.8750 (96.7188) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4284 (1.4331) acc 96.8750 (94.4531) lr 0.260000 -FPS@all 842.902, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:32 loss 1.4260 (1.4160) acc 93.7500 (96.0938) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:23 loss 1.5616 (1.4525) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 842.747, TIME@all 0.304 -epoch: [33/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:20:31 loss 1.2951 (1.3682) acc 96.8750 (97.1875) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4112 (1.4399) acc 93.7500 (94.6875) lr 0.260000 -FPS@all 842.894, TIME@all 0.304 -epoch: [33/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:20:30 loss 1.3132 (1.4014) acc 100.0000 (95.4688) lr 0.260000 -epoch: [33/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:22 loss 1.4165 (1.4387) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 843.127, TIME@all 0.304 -epoch: [34/350][20/50] time 0.305 (0.305) data 0.000 (0.011) eta 1:20:26 loss 1.4535 (1.4110) acc 96.8750 (94.3750) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.6118 (1.4344) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 842.664, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:20:26 loss 1.4155 (1.3867) acc 93.7500 (95.1562) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.5073 (1.4276) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 842.570, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:20:26 loss 1.6079 (1.4238) acc 90.6250 (94.5312) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.3945 (1.4345) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 842.610, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:26 loss 1.2972 (1.3779) acc 96.8750 (96.0938) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.4440 (1.4058) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 842.596, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:26 loss 1.2703 (1.4125) acc 100.0000 (95.4688) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.3298 (1.4398) acc 100.0000 (94.6875) lr 0.260000 -FPS@all 842.610, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:20:25 loss 1.4618 (1.4121) acc 87.5000 (94.8438) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:11 loss 1.7260 (1.4431) acc 84.3750 (94.3750) lr 0.260000 -FPS@all 842.787, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:20:25 loss 1.2716 (1.4210) acc 100.0000 (95.6250) lr 0.260000 -epoch: [34/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:20:11 loss 1.8364 (1.4641) acc 81.2500 (93.6719) lr 0.260000 -FPS@all 842.730, TIME@all 0.304 -epoch: [34/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:20:25 loss 1.3431 (1.3800) acc 93.7500 (96.0938) lr 0.260000 -epoch: [34/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:20:12 loss 1.5976 (1.4302) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 842.996, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.4167 (1.3761) acc 96.8750 (96.0938) lr 0.260000 -epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:51 loss 1.5788 (1.4016) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 843.027, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.5644 (1.3769) acc 90.6250 (95.3125) lr 0.260000 -epoch: [35/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:50 loss 1.4647 (1.4222) acc 100.0000 (94.4531) lr 0.260000 -FPS@all 843.069, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:19:58 loss 1.4607 (1.3885) acc 93.7500 (95.6250) lr 0.260000 -epoch: [35/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 1:19:50 loss 1.6069 (1.4099) acc 93.7500 (94.9219) lr 0.260000 -FPS@all 843.058, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.4376 (1.3777) acc 93.7500 (96.5625) lr 0.260000 -epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:49 loss 1.3996 (1.4042) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 843.254, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:58 loss 1.5132 (1.3841) acc 96.8750 (96.2500) lr 0.260000 -epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:51 loss 1.6493 (1.4118) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 843.050, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.3336 (1.3741) acc 96.8750 (95.6250) lr 0.260000 -epoch: [35/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:50 loss 1.3826 (1.3967) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 843.219, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:57 loss 1.4977 (1.3960) acc 84.3750 (95.3125) lr 0.260000 -epoch: [35/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:50 loss 1.5070 (1.4029) acc 93.7500 (95.0000) lr 0.260000 -FPS@all 843.474, TIME@all 0.304 -epoch: [35/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:58 loss 1.3807 (1.3839) acc 100.0000 (95.3125) lr 0.260000 -epoch: [35/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:19:50 loss 1.4870 (1.4141) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 843.048, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:19:57 loss 1.6046 (1.4545) acc 87.5000 (93.1250) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:41 loss 1.3781 (1.4722) acc 96.8750 (92.5781) lr 0.260000 -FPS@all 842.341, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:19:56 loss 1.4084 (1.4327) acc 100.0000 (95.6250) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.2522 (1.4642) acc 100.0000 (94.4531) lr 0.260000 -FPS@all 842.398, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:19:56 loss 1.8281 (1.4579) acc 84.3750 (94.3750) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:19:41 loss 1.3764 (1.4585) acc 93.7500 (94.0625) lr 0.260000 -FPS@all 842.410, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.5766 (1.4145) acc 90.6250 (95.0000) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.3913 (1.4456) acc 96.8750 (94.1406) lr 0.260000 -FPS@all 842.355, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.6903 (1.3756) acc 81.2500 (96.4062) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.3317 (1.4380) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 842.343, TIME@all 0.304 -epoch: [36/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:19:55 loss 1.4633 (1.4032) acc 87.5000 (95.3125) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:40 loss 1.6214 (1.4449) acc 90.6250 (93.7500) lr 0.260000 -FPS@all 842.536, TIME@all 0.304 -epoch: [36/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:19:54 loss 1.4693 (1.4047) acc 93.7500 (95.6250) lr 0.260000 -epoch: [36/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:19:41 loss 1.7239 (1.4646) acc 87.5000 (93.7500) lr 0.260000 -FPS@all 842.805, TIME@all 0.304 -epoch: [36/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:19:56 loss 1.4292 (1.3865) acc 90.6250 (95.7812) lr 0.260000 -epoch: [36/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:40 loss 1.3963 (1.4119) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 842.485, TIME@all 0.304 -epoch: [37/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.3809 (1.3763) acc 96.8750 (96.0938) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:14 loss 1.5800 (1.4321) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 844.453, TIME@all 0.303 -epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:19:25 loss 1.5023 (1.4105) acc 87.5000 (94.0625) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:19:15 loss 1.3651 (1.4558) acc 96.8750 (93.2812) lr 0.260000 -FPS@all 844.347, TIME@all 0.303 -epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.5963 (1.4039) acc 93.7500 (95.4688) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.3469 (1.4579) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 844.389, TIME@all 0.303 -epoch: [37/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.5471 (1.3827) acc 90.6250 (95.9375) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.6800 (1.4578) acc 87.5000 (93.9844) lr 0.260000 -FPS@all 844.388, TIME@all 0.303 -epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:25 loss 1.3834 (1.3827) acc 93.7500 (95.7812) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:19:15 loss 1.4566 (1.4524) acc 90.6250 (93.4375) lr 0.260000 -FPS@all 844.370, TIME@all 0.303 -epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:24 loss 1.3389 (1.3897) acc 100.0000 (96.4062) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:14 loss 1.3661 (1.4559) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 844.564, TIME@all 0.303 -epoch: [37/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:19:24 loss 1.4975 (1.3775) acc 93.7500 (95.9375) lr 0.260000 -epoch: [37/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:19:14 loss 1.2930 (1.4391) acc 100.0000 (94.3750) lr 0.260000 -FPS@all 844.536, TIME@all 0.303 -epoch: [37/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 1:19:21 loss 1.5539 (1.3801) acc 87.5000 (96.4062) lr 0.260000 -epoch: [37/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:19:15 loss 1.4530 (1.4456) acc 93.7500 (94.8438) lr 0.260000 -FPS@all 844.729, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4846 (1.3458) acc 90.6250 (96.4062) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.3086 (1.3842) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 844.408, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 1:19:01 loss 1.3799 (1.3820) acc 96.8750 (95.3125) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:19:02 loss 1.4939 (1.4265) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 844.299, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4040 (1.3686) acc 96.8750 (96.7188) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.3801 (1.4058) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 844.340, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.3569 (1.3729) acc 96.8750 (96.4062) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.5525 (1.4126) acc 90.6250 (95.0000) lr 0.260000 -FPS@all 844.320, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 1:19:00 loss 1.4394 (1.3431) acc 96.8750 (97.3438) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:01 loss 1.2587 (1.3935) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 844.530, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4167 (1.3556) acc 96.8750 (96.2500) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:02 loss 1.4004 (1.3987) acc 90.6250 (95.2344) lr 0.260000 -FPS@all 844.343, TIME@all 0.303 -epoch: [38/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.5171 (1.3479) acc 90.6250 (96.7188) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:19:01 loss 1.4844 (1.3842) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 844.481, TIME@all 0.303 -epoch: [38/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:19:01 loss 1.4400 (1.3629) acc 90.6250 (95.9375) lr 0.260000 -epoch: [38/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 1:19:01 loss 1.3534 (1.3905) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 844.672, TIME@all 0.303 -epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:19:12 loss 1.3539 (1.3648) acc 96.8750 (96.4062) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:18:57 loss 1.3486 (1.4230) acc 93.7500 (94.6875) lr 0.260000 -FPS@all 842.163, TIME@all 0.304 -epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:19:12 loss 1.4385 (1.3913) acc 96.8750 (95.7812) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:18:56 loss 1.4023 (1.4524) acc 90.6250 (93.9062) lr 0.260000 -FPS@all 842.225, TIME@all 0.304 -epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.3853 (1.3781) acc 96.8750 (95.9375) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:56 loss 1.3193 (1.4065) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 842.234, TIME@all 0.304 -epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.4674 (1.3877) acc 93.7500 (95.6250) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:57 loss 1.3463 (1.4210) acc 93.7500 (94.9219) lr 0.260000 -FPS@all 842.173, TIME@all 0.304 -epoch: [39/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:19:12 loss 1.4147 (1.4315) acc 100.0000 (94.2188) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:57 loss 1.3317 (1.4443) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 842.172, TIME@all 0.304 -epoch: [39/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 1:19:11 loss 1.3255 (1.3796) acc 96.8750 (95.9375) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:55 loss 1.4420 (1.4289) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 842.378, TIME@all 0.304 -epoch: [39/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 1:19:11 loss 1.5030 (1.3760) acc 90.6250 (96.2500) lr 0.260000 -epoch: [39/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:18:56 loss 1.4192 (1.4320) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 842.515, TIME@all 0.304 -epoch: [39/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:19:11 loss 1.5416 (1.4052) acc 96.8750 (96.0938) lr 0.260000 -epoch: [39/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:18:56 loss 1.5239 (1.4596) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 842.323, TIME@all 0.304 -epoch: [40/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:19:00 loss 1.2576 (1.4050) acc 100.0000 (96.2500) lr 0.260000 -epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.4982 (1.4261) acc 90.6250 (95.2344) lr 0.260000 -FPS@all 840.898, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:18:59 loss 1.3215 (1.3954) acc 100.0000 (95.6250) lr 0.260000 -epoch: [40/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.5421 (1.4280) acc 90.6250 (94.6875) lr 0.260000 -FPS@all 840.967, TIME@all 0.304 -epoch: [40/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 1:18:59 loss 1.3232 (1.3708) acc 100.0000 (96.5625) lr 0.260000 -epoch: [40/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.4798 (1.4092) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 840.975, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:19:00 loss 1.4074 (1.3683) acc 93.7500 (95.6250) lr 0.260000 -epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:18:48 loss 1.4727 (1.4155) acc 90.6250 (94.6094) lr 0.260000 -FPS@all 840.922, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.012) eta 1:19:00 loss 1.3298 (1.3658) acc 96.8750 (96.4062) lr 0.260000 -epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 1:18:48 loss 1.3420 (1.4125) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 840.918, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:18:59 loss 1.3986 (1.3764) acc 93.7500 (96.4062) lr 0.260000 -epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 1:18:47 loss 1.5653 (1.3993) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 841.058, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.013) eta 1:18:58 loss 1.3725 (1.3840) acc 93.7500 (95.6250) lr 0.260000 -epoch: [40/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 1:18:47 loss 1.5646 (1.4220) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 841.118, TIME@all 0.304 -epoch: [40/350][20/50] time 0.308 (0.305) data 0.001 (0.012) eta 1:19:00 loss 1.3692 (1.4003) acc 96.8750 (95.9375) lr 0.260000 -epoch: [40/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:18:48 loss 1.6419 (1.4250) acc 90.6250 (94.7656) lr 0.260000 -FPS@all 841.230, TIME@all 0.304 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:18:26 loss 1.4732 (1.4418) acc 93.7500 (95.6250) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.4219 (1.4474) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 843.876, TIME@all 0.303 -epoch: [41/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:18:27 loss 1.5240 (1.4289) acc 93.7500 (94.2188) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.4799 (1.4378) acc 90.6250 (94.2969) lr 0.260000 -FPS@all 843.845, TIME@all 0.303 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:18:26 loss 1.4274 (1.4163) acc 100.0000 (94.2188) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.5713 (1.4443) acc 90.6250 (93.9844) lr 0.260000 -FPS@all 843.865, TIME@all 0.303 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:26 loss 1.5023 (1.4226) acc 90.6250 (94.5312) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:19 loss 1.4060 (1.4299) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 844.034, TIME@all 0.303 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:27 loss 1.6324 (1.4363) acc 90.6250 (95.1562) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:20 loss 1.7515 (1.4644) acc 87.5000 (93.4375) lr 0.260000 -FPS@all 843.850, TIME@all 0.303 -epoch: [41/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:18:25 loss 1.7913 (1.4328) acc 81.2500 (93.7500) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:18:20 loss 1.3862 (1.4307) acc 96.8750 (94.1406) lr 0.260000 -FPS@all 844.204, TIME@all 0.303 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:27 loss 1.3687 (1.4201) acc 96.8750 (94.8438) lr 0.260000 -epoch: [41/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:18:20 loss 1.4505 (1.4408) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 843.834, TIME@all 0.303 -epoch: [41/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:18:26 loss 1.3919 (1.4265) acc 96.8750 (95.0000) lr 0.260000 -epoch: [41/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:18:19 loss 1.3551 (1.4397) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 844.001, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.3690 (1.3649) acc 96.8750 (95.7812) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:59 loss 1.3367 (1.3923) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 844.188, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.7243 (1.4070) acc 87.5000 (95.6250) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:59 loss 1.3382 (1.4009) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 844.224, TIME@all 0.303 -epoch: [42/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 1:17:52 loss 1.5320 (1.3692) acc 96.8750 (96.5625) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:59 loss 1.3697 (1.4028) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 844.194, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.4798 (1.4165) acc 93.7500 (95.3125) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.2726 (1.4100) acc 100.0000 (94.9219) lr 0.260000 -FPS@all 844.377, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:50 loss 1.3823 (1.3713) acc 100.0000 (95.9375) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:17:58 loss 1.2971 (1.3939) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 844.623, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.5232 (1.3893) acc 87.5000 (94.6875) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.3591 (1.3996) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 844.241, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:52 loss 1.4646 (1.3977) acc 93.7500 (95.6250) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.2645 (1.3895) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 844.231, TIME@all 0.303 -epoch: [42/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:17:51 loss 1.5353 (1.3767) acc 93.7500 (95.3125) lr 0.260000 -epoch: [42/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:17:58 loss 1.4745 (1.4077) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 844.415, TIME@all 0.303 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.3505 (1.3229) acc 93.7500 (96.8750) lr 0.260000 -epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.4131 (1.3743) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 842.503, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.4861 (1.3385) acc 90.6250 (96.7188) lr 0.260000 -epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:54 loss 1.7041 (1.3833) acc 87.5000 (95.7812) lr 0.260000 -FPS@all 842.432, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.6326 (1.3280) acc 84.3750 (96.2500) lr 0.260000 -epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:53 loss 1.5188 (1.3630) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 842.484, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:18:04 loss 1.3970 (1.3364) acc 96.8750 (96.5625) lr 0.260000 -epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.4296 (1.3693) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 842.507, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.3655 (1.2987) acc 96.8750 (97.5000) lr 0.260000 -epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:54 loss 1.4523 (1.3554) acc 90.6250 (96.0156) lr 0.260000 -FPS@all 842.439, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:18:03 loss 1.4184 (1.3275) acc 93.7500 (97.0312) lr 0.260000 -epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:17:52 loss 1.3124 (1.3582) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 842.666, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:03 loss 1.5042 (1.3457) acc 90.6250 (97.5000) lr 0.260000 -epoch: [43/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:17:53 loss 1.6956 (1.3947) acc 78.1250 (95.8594) lr 0.260000 -FPS@all 842.614, TIME@all 0.304 -epoch: [43/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:18:04 loss 1.4499 (1.3134) acc 96.8750 (97.3438) lr 0.260000 -epoch: [43/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:17:53 loss 1.2931 (1.3483) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 842.764, TIME@all 0.304 -epoch: [44/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:17:45 loss 1.5203 (1.3899) acc 90.6250 (95.9375) lr 0.260000 -epoch: [44/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.3939 (1.4193) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 842.873, TIME@all 0.304 -epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:17:47 loss 1.4746 (1.3596) acc 87.5000 (96.2500) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.4934 (1.3994) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 842.814, TIME@all 0.304 -epoch: [44/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:17:46 loss 1.4012 (1.3652) acc 93.7500 (96.2500) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.2976 (1.3924) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 842.871, TIME@all 0.304 -epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:17:46 loss 1.3537 (1.3492) acc 93.7500 (96.2500) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:17:40 loss 1.2906 (1.3904) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 842.852, TIME@all 0.304 -epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:17:46 loss 1.2827 (1.3777) acc 96.8750 (96.0938) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:17:40 loss 1.4168 (1.4054) acc 96.8750 (94.9219) lr 0.260000 -FPS@all 842.816, TIME@all 0.304 -epoch: [44/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:17:45 loss 1.3694 (1.3779) acc 96.8750 (96.2500) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:17:39 loss 1.2714 (1.3988) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 843.001, TIME@all 0.304 -epoch: [44/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:17:46 loss 1.3043 (1.3765) acc 100.0000 (95.6250) lr 0.260000 -epoch: [44/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:17:39 loss 1.4881 (1.4039) acc 87.5000 (94.9219) lr 0.260000 -FPS@all 842.952, TIME@all 0.304 -epoch: [44/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:17:47 loss 1.4765 (1.3700) acc 93.7500 (95.7812) lr 0.260000 -epoch: [44/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:17:40 loss 1.4077 (1.4033) acc 93.7500 (95.2344) lr 0.260000 -FPS@all 843.157, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:17:42 loss 1.2758 (1.3500) acc 100.0000 (96.8750) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:17:31 loss 1.3525 (1.3929) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 841.247, TIME@all 0.304 -epoch: [45/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:17:42 loss 1.5576 (1.3549) acc 93.7500 (97.1875) lr 0.260000 -epoch: [45/350][40/50] time 0.306 (0.305) data 0.001 (0.006) eta 1:17:30 loss 1.4492 (1.4208) acc 93.7500 (94.6875) lr 0.260000 -FPS@all 841.299, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:17:41 loss 1.5842 (1.3539) acc 87.5000 (96.0938) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.3953 (1.3915) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 841.291, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:17:41 loss 1.3760 (1.3457) acc 96.8750 (97.0312) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:31 loss 1.4133 (1.3739) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 841.280, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:17:41 loss 1.3082 (1.3692) acc 100.0000 (95.7812) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.4138 (1.3949) acc 96.8750 (94.9219) lr 0.260000 -FPS@all 841.394, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 1:17:42 loss 1.3156 (1.3498) acc 96.8750 (96.8750) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:31 loss 1.3889 (1.3869) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 841.214, TIME@all 0.304 -epoch: [45/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 1:17:41 loss 1.5638 (1.3643) acc 90.6250 (95.4688) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.5100 (1.4101) acc 87.5000 (94.2969) lr 0.260000 -FPS@all 841.456, TIME@all 0.304 -epoch: [45/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:17:38 loss 1.3572 (1.3411) acc 96.8750 (96.7188) lr 0.260000 -epoch: [45/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:17:30 loss 1.3643 (1.3695) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 841.684, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:17:23 loss 1.6539 (1.4238) acc 90.6250 (94.5312) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.006) eta 1:17:15 loss 1.3770 (1.4462) acc 96.8750 (93.9844) lr 0.260000 -FPS@all 841.317, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:17:23 loss 1.4017 (1.3992) acc 96.8750 (95.4688) lr 0.260000 -epoch: [46/350][40/50] time 0.309 (0.305) data 0.000 (0.006) eta 1:17:15 loss 1.3529 (1.4435) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 841.363, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.5369 (1.3698) acc 87.5000 (96.2500) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:15 loss 1.4765 (1.4204) acc 96.8750 (94.8438) lr 0.260000 -FPS@all 841.342, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.3305 (1.3616) acc 100.0000 (96.2500) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:15 loss 1.3361 (1.3974) acc 96.8750 (94.9219) lr 0.260000 -FPS@all 841.351, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:17:22 loss 1.7191 (1.3910) acc 90.6250 (95.4688) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.001 (0.007) eta 1:17:14 loss 1.3428 (1.4436) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 841.541, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.6737 (1.3984) acc 84.3750 (95.1562) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.001 (0.007) eta 1:17:15 loss 1.4671 (1.4560) acc 90.6250 (93.2031) lr 0.260000 -FPS@all 841.319, TIME@all 0.304 -epoch: [46/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:17:22 loss 1.5724 (1.3709) acc 90.6250 (96.4062) lr 0.260000 -epoch: [46/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:17:14 loss 1.4607 (1.4243) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 841.702, TIME@all 0.304 -epoch: [46/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:17:23 loss 1.4585 (1.3717) acc 93.7500 (96.8750) lr 0.260000 -epoch: [46/350][40/50] time 0.310 (0.305) data 0.000 (0.007) eta 1:17:14 loss 1.5116 (1.4253) acc 90.6250 (94.6875) lr 0.260000 -FPS@all 841.459, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:16:55 loss 1.4941 (1.3669) acc 90.6250 (96.7188) lr 0.260000 -epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:16:52 loss 1.3977 (1.4263) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 842.742, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:56 loss 1.6213 (1.3811) acc 84.3750 (95.6250) lr 0.260000 -epoch: [47/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.2942 (1.4394) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 842.745, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:16:55 loss 1.5742 (1.3783) acc 87.5000 (96.4062) lr 0.260000 -epoch: [47/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:16:52 loss 1.4398 (1.4173) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 842.669, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:16:54 loss 1.7294 (1.3950) acc 81.2500 (95.9375) lr 0.260000 -epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:51 loss 1.5283 (1.4531) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 842.887, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 1:16:55 loss 1.6822 (1.3763) acc 87.5000 (95.6250) lr 0.260000 -epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:51 loss 1.4442 (1.4378) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 842.833, TIME@all 0.304 -epoch: [47/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:54 loss 1.5721 (1.3950) acc 93.7500 (95.4688) lr 0.260000 -epoch: [47/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 1:16:51 loss 1.3653 (1.4449) acc 96.8750 (93.7500) lr 0.260000 -FPS@all 843.030, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:55 loss 1.6564 (1.4090) acc 90.6250 (95.0000) lr 0.260000 -epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.3278 (1.4365) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 842.694, TIME@all 0.304 -epoch: [47/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:55 loss 1.4687 (1.4010) acc 100.0000 (95.0000) lr 0.260000 -epoch: [47/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:52 loss 1.3657 (1.4330) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 842.690, TIME@all 0.304 -epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:16:43 loss 1.6637 (1.3769) acc 90.6250 (96.0938) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:16:34 loss 1.4681 (1.4196) acc 96.8750 (94.6094) lr 0.260000 -FPS@all 842.709, TIME@all 0.304 -epoch: [48/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:16:43 loss 1.3808 (1.3219) acc 100.0000 (96.5625) lr 0.260000 -epoch: [48/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.4481 (1.3912) acc 87.5000 (94.2969) lr 0.260000 -FPS@all 842.756, TIME@all 0.304 -epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:44 loss 1.5695 (1.3755) acc 96.8750 (97.1875) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.3970 (1.4171) acc 93.7500 (94.8438) lr 0.260000 -FPS@all 842.706, TIME@all 0.304 -epoch: [48/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:16:44 loss 1.5247 (1.3576) acc 90.6250 (96.5625) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.5772 (1.4216) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 842.750, TIME@all 0.304 -epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:42 loss 1.5281 (1.3592) acc 90.6250 (95.9375) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.4027 (1.4067) acc 100.0000 (94.9219) lr 0.260000 -FPS@all 842.941, TIME@all 0.304 -epoch: [48/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:16:43 loss 1.6272 (1.3828) acc 90.6250 (95.9375) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 1:16:33 loss 1.4028 (1.4183) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 843.062, TIME@all 0.304 -epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:43 loss 1.8314 (1.3917) acc 78.1250 (95.0000) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:33 loss 1.5794 (1.4468) acc 81.2500 (93.5156) lr 0.260000 -FPS@all 842.866, TIME@all 0.304 -epoch: [48/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:16:44 loss 1.4955 (1.3893) acc 93.7500 (96.4062) lr 0.260000 -epoch: [48/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 1:16:34 loss 1.2927 (1.4312) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 842.708, TIME@all 0.304 -epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:16:20 loss 1.4345 (1.3489) acc 96.8750 (96.2500) lr 0.260000 -epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:16:20 loss 1.5156 (1.3939) acc 90.6250 (95.1562) lr 0.260000 -FPS@all 843.088, TIME@all 0.304 -epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.3402 (1.3384) acc 100.0000 (98.1250) lr 0.260000 -epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.4750 (1.3923) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 843.120, TIME@all 0.304 -epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.3404 (1.3462) acc 96.8750 (96.8750) lr 0.260000 -epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.2914 (1.3886) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 843.101, TIME@all 0.304 -epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.4825 (1.3539) acc 93.7500 (96.4062) lr 0.260000 -epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.5743 (1.3934) acc 90.6250 (95.3125) lr 0.260000 -FPS@all 843.302, TIME@all 0.304 -epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.3827 (1.3453) acc 93.7500 (96.5625) lr 0.260000 -epoch: [49/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.3679 (1.3691) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 843.108, TIME@all 0.304 -epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:20 loss 1.5449 (1.3847) acc 90.6250 (95.4688) lr 0.260000 -epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.4562 (1.4327) acc 96.8750 (93.7500) lr 0.260000 -FPS@all 843.095, TIME@all 0.304 -epoch: [49/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:16:19 loss 1.4416 (1.3709) acc 96.8750 (96.0938) lr 0.260000 -epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:19 loss 1.6143 (1.4155) acc 90.6250 (94.2969) lr 0.260000 -FPS@all 843.216, TIME@all 0.304 -epoch: [49/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:16:21 loss 1.4899 (1.3760) acc 90.6250 (95.1562) lr 0.260000 -epoch: [49/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:16:20 loss 1.4687 (1.3896) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 843.407, TIME@all 0.304 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4668 (1.3481) acc 90.6250 (96.4062) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.2377 (1.3754) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 843.535, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4682 (1.3419) acc 93.7500 (97.8125) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.3581 (1.3561) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 843.624, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.3858 (1.3712) acc 90.6250 (96.4062) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.2464 (1.3669) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 843.573, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.3636 (1.3229) acc 100.0000 (97.6562) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.4016 (1.3488) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 843.543, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:14 loss 1.4267 (1.3426) acc 93.7500 (96.7188) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:16:00 loss 1.5222 (1.3618) acc 90.6250 (96.0156) lr 0.260000 -FPS@all 843.533, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:16:13 loss 1.3286 (1.3620) acc 100.0000 (96.5625) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 1:15:58 loss 1.2813 (1.3754) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 843.758, TIME@all 0.303 -epoch: [50/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:16:13 loss 1.3820 (1.3516) acc 100.0000 (96.0938) lr 0.260000 -epoch: [50/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.2784 (1.3839) acc 100.0000 (95.4688) lr 0.260000 -FPS@all 843.680, TIME@all 0.303 -epoch: [50/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:16:15 loss 1.6284 (1.3816) acc 87.5000 (95.4688) lr 0.260000 -epoch: [50/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:15:59 loss 1.3628 (1.4008) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 843.836, TIME@all 0.303 -epoch: [51/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:16:08 loss 1.4294 (1.3719) acc 93.7500 (95.7812) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.3863 (1.3848) acc 90.6250 (95.3125) lr 0.260000 -FPS@all 842.879, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.4116 (1.3406) acc 96.8750 (95.9375) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.4322 (1.3785) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 842.831, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.3801 (1.3339) acc 100.0000 (97.5000) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.001 (0.006) eta 1:15:51 loss 1.4403 (1.3734) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 842.893, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.4574 (1.3464) acc 96.8750 (97.3438) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.4311 (1.3841) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 842.874, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:16:09 loss 1.5432 (1.3577) acc 87.5000 (96.5625) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:15:51 loss 1.5316 (1.3790) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 842.876, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.3982 (1.3600) acc 100.0000 (96.0938) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:15:50 loss 1.4276 (1.3885) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 843.062, TIME@all 0.304 -epoch: [51/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:16:08 loss 1.5042 (1.3501) acc 93.7500 (95.7812) lr 0.260000 -epoch: [51/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 1:15:50 loss 1.3474 (1.3859) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 843.283, TIME@all 0.304 -epoch: [51/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:16:08 loss 1.2923 (1.3303) acc 100.0000 (97.3438) lr 0.260000 -epoch: [51/350][40/50] time 0.297 (0.304) data 0.001 (0.007) eta 1:15:50 loss 1.5538 (1.3658) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 842.996, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.5025 (1.3559) acc 90.6250 (96.2500) lr 0.260000 -epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:31 loss 1.4122 (1.3741) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 843.277, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:42 loss 1.3861 (1.3450) acc 93.7500 (95.9375) lr 0.260000 -epoch: [52/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4924 (1.3935) acc 93.7500 (95.0000) lr 0.260000 -FPS@all 843.326, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.3561 (1.3438) acc 96.8750 (95.7812) lr 0.260000 -epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:30 loss 1.3935 (1.3660) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 843.389, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4355 (1.3560) acc 93.7500 (96.2500) lr 0.260000 -epoch: [52/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4628 (1.3879) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 843.328, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:42 loss 1.4102 (1.3808) acc 93.7500 (95.7812) lr 0.260000 -epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:15:30 loss 1.4888 (1.3926) acc 90.6250 (95.1562) lr 0.260000 -FPS@all 843.300, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4135 (1.3505) acc 93.7500 (95.6250) lr 0.260000 -epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:15:29 loss 1.5182 (1.3806) acc 87.5000 (95.2344) lr 0.260000 -FPS@all 843.485, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:41 loss 1.4932 (1.3464) acc 90.6250 (96.8750) lr 0.260000 -epoch: [52/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:15:30 loss 1.4908 (1.3912) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 843.443, TIME@all 0.304 -epoch: [52/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:15:41 loss 1.5703 (1.3512) acc 87.5000 (96.2500) lr 0.260000 -epoch: [52/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:15:29 loss 1.4527 (1.3794) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 843.610, TIME@all 0.303 -epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:15:31 loss 1.2600 (1.3451) acc 100.0000 (96.2500) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.4386 (1.3544) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 843.322, TIME@all 0.304 -epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.2617 (1.3504) acc 100.0000 (96.7188) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:16 loss 1.3262 (1.3588) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 843.428, TIME@all 0.304 -epoch: [53/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3459 (1.3519) acc 96.8750 (96.7188) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:16 loss 1.3717 (1.3731) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 843.365, TIME@all 0.304 -epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:32 loss 1.3196 (1.3398) acc 96.8750 (96.8750) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.5282 (1.3740) acc 90.6250 (96.0156) lr 0.260000 -FPS@all 843.346, TIME@all 0.304 -epoch: [53/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:15:30 loss 1.4106 (1.3634) acc 93.7500 (95.3125) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.007) eta 1:15:16 loss 1.3932 (1.3710) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 843.546, TIME@all 0.303 -epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3302 (1.3481) acc 96.8750 (96.2500) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.3857 (1.3613) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 843.331, TIME@all 0.304 -epoch: [53/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:15:31 loss 1.3498 (1.3419) acc 93.7500 (96.8750) lr 0.260000 -epoch: [53/350][40/50] time 0.298 (0.304) data 0.000 (0.007) eta 1:15:16 loss 1.5431 (1.3683) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 843.480, TIME@all 0.304 -epoch: [53/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:15:31 loss 1.3679 (1.3353) acc 93.7500 (96.8750) lr 0.260000 -epoch: [53/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 1:15:17 loss 1.3967 (1.3707) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 843.681, TIME@all 0.303 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:15:10 loss 1.3664 (1.3004) acc 96.8750 (97.9688) lr 0.260000 -epoch: [54/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.2155 (1.3411) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 842.744, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:15:10 loss 1.2573 (1.3227) acc 100.0000 (96.5625) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.2661 (1.3564) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 842.745, TIME@all 0.304 -epoch: [54/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 1:15:11 loss 1.3846 (1.3395) acc 96.8750 (97.0312) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:02 loss 1.2787 (1.3624) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 842.665, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:10 loss 1.3902 (1.3296) acc 93.7500 (97.1875) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:02 loss 1.3812 (1.3678) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 842.692, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:11 loss 1.3139 (1.3080) acc 96.8750 (97.3438) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:15:02 loss 1.3454 (1.3469) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 842.678, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:10 loss 1.2916 (1.3188) acc 96.8750 (96.5625) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:01 loss 1.4392 (1.3537) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 842.844, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:15:13 loss 1.4681 (1.3321) acc 90.6250 (97.5000) lr 0.260000 -epoch: [54/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:15:04 loss 1.3051 (1.3581) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 842.874, TIME@all 0.304 -epoch: [54/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 1:15:10 loss 1.2272 (1.3215) acc 100.0000 (97.9688) lr 0.260000 -epoch: [54/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:15:01 loss 1.3147 (1.3649) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 842.880, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:09 loss 1.3413 (1.3454) acc 93.7500 (95.7812) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.6728 (1.3724) acc 90.6250 (95.2344) lr 0.260000 -FPS@all 841.664, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:09 loss 1.4178 (1.3512) acc 96.8750 (96.2500) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.4615 (1.3747) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 841.718, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:15:07 loss 1.5324 (1.3393) acc 93.7500 (96.8750) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:57 loss 1.4710 (1.3502) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 841.783, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 1:15:09 loss 1.4522 (1.3468) acc 90.6250 (96.7188) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:58 loss 1.4574 (1.3604) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 841.695, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:09 loss 1.7154 (1.3533) acc 90.6250 (97.1875) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:58 loss 1.4411 (1.3743) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 841.646, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:08 loss 1.4800 (1.3501) acc 93.7500 (96.7188) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.4767 (1.3454) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 841.810, TIME@all 0.304 -epoch: [55/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:15:08 loss 1.5136 (1.3481) acc 90.6250 (96.5625) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.5412 (1.3524) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 841.861, TIME@all 0.304 -epoch: [55/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:15:07 loss 1.5582 (1.3241) acc 96.8750 (97.1875) lr 0.260000 -epoch: [55/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:14:57 loss 1.3947 (1.3474) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 842.116, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2974 (1.3412) acc 100.0000 (96.2500) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.2866 (1.3650) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 842.501, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:48 loss 1.3311 (1.3352) acc 93.7500 (96.7188) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4252 (1.3733) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 842.520, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2290 (1.3495) acc 100.0000 (95.7812) lr 0.260000 -epoch: [56/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4373 (1.3817) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 842.407, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2738 (1.3499) acc 100.0000 (97.1875) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.3066 (1.3605) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 842.468, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:14:48 loss 1.2391 (1.3109) acc 100.0000 (97.6562) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:38 loss 1.3469 (1.3458) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 842.645, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.2903 (1.3422) acc 96.8750 (96.8750) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.4900 (1.3550) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 842.471, TIME@all 0.304 -epoch: [56/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:14:49 loss 1.2782 (1.3303) acc 100.0000 (96.8750) lr 0.260000 -epoch: [56/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:38 loss 1.3370 (1.3529) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 842.579, TIME@all 0.304 -epoch: [56/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:14:49 loss 1.3400 (1.3395) acc 96.8750 (96.4062) lr 0.260000 -epoch: [56/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:14:39 loss 1.5472 (1.3847) acc 90.6250 (95.2344) lr 0.260000 -FPS@all 842.729, TIME@all 0.304 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.011) eta 1:14:18 loss 1.2647 (1.3287) acc 100.0000 (97.3438) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.3021 (1.3641) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 845.171, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 1:14:17 loss 1.4495 (1.3449) acc 93.7500 (95.7812) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:04 loss 1.4306 (1.3709) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 845.250, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:14:18 loss 1.4836 (1.3410) acc 90.6250 (95.9375) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 1:14:05 loss 1.2225 (1.3727) acc 100.0000 (95.3906) lr 0.260000 -FPS@all 845.195, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.3485 (1.3326) acc 96.8750 (97.0312) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.4996 (1.3601) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 845.188, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.4348 (1.3139) acc 90.6250 (96.8750) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 1:14:05 loss 1.5361 (1.3592) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 845.190, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:14:16 loss 1.2825 (1.3057) acc 100.0000 (97.1875) lr 0.260000 -epoch: [57/350][40/50] time 0.300 (0.303) data 0.001 (0.007) eta 1:14:04 loss 1.3572 (1.3613) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 845.385, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 1:14:16 loss 1.3932 (1.3411) acc 100.0000 (97.0312) lr 0.260000 -epoch: [57/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 1:14:04 loss 1.3425 (1.3682) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 845.356, TIME@all 0.303 -epoch: [57/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 1:14:17 loss 1.4809 (1.3093) acc 87.5000 (97.0312) lr 0.260000 -epoch: [57/350][40/50] time 0.303 (0.303) data 0.001 (0.006) eta 1:14:04 loss 1.4408 (1.3593) acc 93.7500 (95.8594) lr 0.260000 -FPS@all 845.540, TIME@all 0.303 -epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.4080 (1.3206) acc 100.0000 (96.7188) lr 0.260000 -epoch: [58/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.2940 (1.3406) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 842.820, TIME@all 0.304 -epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.2434 (1.3120) acc 100.0000 (97.1875) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.3424 (1.3531) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 842.803, TIME@all 0.304 -epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 1:14:22 loss 1.4265 (1.3244) acc 96.8750 (97.0312) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:14:07 loss 1.3450 (1.3476) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 842.742, TIME@all 0.304 -epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.014) eta 1:14:21 loss 1.3292 (1.3288) acc 96.8750 (95.9375) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.4119 (1.3536) acc 93.7500 (95.7812) lr 0.260000 -FPS@all 842.915, TIME@all 0.304 -epoch: [58/350][20/50] time 0.309 (0.305) data 0.000 (0.014) eta 1:14:20 loss 1.3882 (1.3274) acc 90.6250 (96.5625) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:06 loss 1.3441 (1.3380) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 842.962, TIME@all 0.304 -epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:21 loss 1.4081 (1.3096) acc 96.8750 (97.0312) lr 0.260000 -epoch: [58/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.3607 (1.3365) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 842.779, TIME@all 0.304 -epoch: [58/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 1:14:22 loss 1.3045 (1.3249) acc 100.0000 (96.8750) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:14:07 loss 1.2861 (1.3527) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 842.754, TIME@all 0.304 -epoch: [58/350][20/50] time 0.309 (0.305) data 0.001 (0.013) eta 1:14:20 loss 1.3346 (1.3157) acc 96.8750 (96.4062) lr 0.260000 -epoch: [58/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:14:06 loss 1.2801 (1.3490) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 843.163, TIME@all 0.304 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4228 (1.3389) acc 93.7500 (97.3438) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3923 (1.3584) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 843.461, TIME@all 0.304 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4597 (1.3366) acc 96.8750 (97.0312) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.2432 (1.3569) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 843.422, TIME@all 0.304 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4448 (1.3521) acc 96.8750 (96.8750) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3688 (1.3622) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 843.359, TIME@all 0.304 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:13:57 loss 1.4144 (1.3251) acc 96.8750 (96.7188) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.3814 (1.3450) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 843.393, TIME@all 0.304 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:13:56 loss 1.4230 (1.3192) acc 100.0000 (97.0312) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:13:44 loss 1.4836 (1.3705) acc 90.6250 (95.8594) lr 0.260000 -FPS@all 843.551, TIME@all 0.303 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:13:56 loss 1.4001 (1.3330) acc 93.7500 (96.8750) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:13:44 loss 1.4199 (1.3518) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 843.539, TIME@all 0.303 -epoch: [59/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:56 loss 1.4002 (1.3248) acc 93.7500 (97.1875) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:13:45 loss 1.2870 (1.3363) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 843.732, TIME@all 0.303 -epoch: [59/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:13:57 loss 1.4090 (1.3478) acc 100.0000 (97.1875) lr 0.260000 -epoch: [59/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:13:45 loss 1.4894 (1.3630) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 843.397, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.3193 (1.3508) acc 96.8750 (96.0938) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.2403 (1.3592) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 843.250, TIME@all 0.304 -epoch: [60/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 1:13:29 loss 1.4471 (1.3171) acc 93.7500 (97.1875) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:13:32 loss 1.2966 (1.3417) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 843.338, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:13:30 loss 1.2352 (1.3504) acc 100.0000 (96.2500) lr 0.260000 -epoch: [60/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:13:33 loss 1.2901 (1.3574) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 843.202, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.3026 (1.3251) acc 100.0000 (97.1875) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.3674 (1.3429) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 843.215, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.2748 (1.3512) acc 96.8750 (96.4062) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.4112 (1.3705) acc 90.6250 (95.5469) lr 0.260000 -FPS@all 843.378, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:30 loss 1.4601 (1.3453) acc 90.6250 (96.5625) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:33 loss 1.5230 (1.3682) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 843.217, TIME@all 0.304 -epoch: [60/350][20/50] time 0.305 (0.303) data 0.000 (0.014) eta 1:13:29 loss 1.2677 (1.3572) acc 96.8750 (96.0938) lr 0.260000 -epoch: [60/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.3091 (1.3582) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 843.422, TIME@all 0.304 -epoch: [60/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 1:13:29 loss 1.3276 (1.3614) acc 93.7500 (95.3125) lr 0.260000 -epoch: [60/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:13:32 loss 1.2973 (1.3666) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 843.675, TIME@all 0.303 -epoch: [61/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2977 (1.2904) acc 93.7500 (97.0312) lr 0.260000 -epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.4225 (1.3346) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 843.274, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:25 loss 1.3211 (1.3150) acc 93.7500 (97.5000) lr 0.260000 -epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.2626 (1.3450) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 843.215, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2201 (1.3379) acc 100.0000 (96.8750) lr 0.260000 -epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:17 loss 1.3599 (1.3795) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 843.248, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2702 (1.2961) acc 96.8750 (96.8750) lr 0.260000 -epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:17 loss 1.2461 (1.3210) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 843.233, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:24 loss 1.3587 (1.3580) acc 100.0000 (96.4062) lr 0.260000 -epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:16 loss 1.3215 (1.3481) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 843.430, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.1878 (1.3174) acc 100.0000 (97.5000) lr 0.260000 -epoch: [61/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:16 loss 1.2605 (1.3268) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 843.375, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:25 loss 1.2797 (1.3195) acc 100.0000 (96.7188) lr 0.260000 -epoch: [61/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.3912 (1.3417) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 843.234, TIME@all 0.304 -epoch: [61/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:13:25 loss 1.2732 (1.2843) acc 100.0000 (98.5938) lr 0.260000 -epoch: [61/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:13:17 loss 1.3154 (1.3203) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 843.593, TIME@all 0.303 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:13:12 loss 1.6327 (1.3188) acc 90.6250 (96.7188) lr 0.260000 -epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:13:06 loss 1.3067 (1.3650) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 842.696, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:13:13 loss 1.5963 (1.3694) acc 87.5000 (95.0000) lr 0.260000 -epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:13:07 loss 1.3803 (1.3873) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 842.613, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.7135 (1.3618) acc 87.5000 (96.0938) lr 0.260000 -epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:13:06 loss 1.4224 (1.4043) acc 96.8750 (95.0781) lr 0.260000 -FPS@all 842.719, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:12 loss 1.4728 (1.3520) acc 93.7500 (95.7812) lr 0.260000 -epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.4655 (1.3760) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 842.800, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.6463 (1.3574) acc 93.7500 (95.7812) lr 0.260000 -epoch: [62/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.4464 (1.3894) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 842.668, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:13:13 loss 1.8071 (1.3706) acc 81.2500 (95.6250) lr 0.260000 -epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.5875 (1.3931) acc 87.5000 (95.0000) lr 0.260000 -FPS@all 842.668, TIME@all 0.304 -epoch: [62/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:13:12 loss 1.5690 (1.3385) acc 84.3750 (96.8750) lr 0.260000 -epoch: [62/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:13:05 loss 1.4180 (1.3784) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 842.823, TIME@all 0.304 -epoch: [62/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:13:10 loss 1.5478 (1.3385) acc 90.6250 (96.8750) lr 0.260000 -epoch: [62/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:13:06 loss 1.3710 (1.3834) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 842.992, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 1:13:02 loss 1.6306 (1.4578) acc 90.6250 (93.5938) lr 0.260000 -epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.3748 (1.4955) acc 96.8750 (92.9688) lr 0.260000 -FPS@all 842.685, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.5835 (1.4247) acc 90.6250 (94.8438) lr 0.260000 -epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.7099 (1.4941) acc 84.3750 (92.7344) lr 0.260000 -FPS@all 842.736, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.7069 (1.4218) acc 87.5000 (95.1562) lr 0.260000 -epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.3751 (1.5016) acc 96.8750 (92.4219) lr 0.260000 -FPS@all 842.721, TIME@all 0.304 -epoch: [63/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 1:13:01 loss 1.6981 (1.4126) acc 90.6250 (95.7812) lr 0.260000 -epoch: [63/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:12:49 loss 1.4541 (1.4850) acc 93.7500 (93.2031) lr 0.260000 -FPS@all 842.747, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:02 loss 1.6007 (1.4060) acc 90.6250 (96.2500) lr 0.260000 -epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.4705 (1.4874) acc 93.7500 (93.8281) lr 0.260000 -FPS@all 842.696, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:01 loss 1.4483 (1.4269) acc 96.8750 (95.4688) lr 0.260000 -epoch: [63/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:12:49 loss 1.5096 (1.4915) acc 93.7500 (94.0625) lr 0.260000 -FPS@all 842.820, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:13:01 loss 1.6165 (1.4250) acc 93.7500 (95.3125) lr 0.260000 -epoch: [63/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:50 loss 1.4450 (1.4697) acc 93.7500 (94.0625) lr 0.260000 -FPS@all 843.040, TIME@all 0.304 -epoch: [63/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:13:01 loss 1.5048 (1.4109) acc 96.8750 (94.6875) lr 0.260000 -epoch: [63/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:12:49 loss 1.5889 (1.4873) acc 93.7500 (92.7344) lr 0.260000 -FPS@all 842.879, TIME@all 0.304 -epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:12:31 loss 1.8600 (1.4437) acc 81.2500 (94.2188) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:12:32 loss 1.3344 (1.4269) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 842.925, TIME@all 0.304 -epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.7978 (1.4431) acc 81.2500 (93.5938) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3815 (1.4709) acc 96.8750 (93.0469) lr 0.260000 -FPS@all 842.982, TIME@all 0.304 -epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.6890 (1.4357) acc 87.5000 (94.3750) lr 0.260000 -epoch: [64/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3250 (1.4359) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 842.971, TIME@all 0.304 -epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 2.0128 (1.4416) acc 71.8750 (94.5312) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.3284 (1.4604) acc 96.8750 (93.5938) lr 0.260000 -FPS@all 842.925, TIME@all 0.304 -epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.7298 (1.4001) acc 90.6250 (95.6250) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:31 loss 1.4155 (1.4284) acc 100.0000 (95.0000) lr 0.260000 -FPS@all 843.070, TIME@all 0.304 -epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:12:30 loss 1.6356 (1.4208) acc 90.6250 (94.5312) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:31 loss 1.3680 (1.4530) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 843.118, TIME@all 0.304 -epoch: [64/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:12:29 loss 1.6852 (1.4209) acc 90.6250 (95.1562) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:12:31 loss 1.5108 (1.4438) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 843.267, TIME@all 0.304 -epoch: [64/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:12:30 loss 2.0156 (1.4235) acc 78.1250 (95.3125) lr 0.260000 -epoch: [64/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:12:32 loss 1.4068 (1.4408) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 842.934, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:17 loss 1.4271 (1.3436) acc 93.7500 (97.8125) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.3954 (1.3674) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 843.054, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:18 loss 1.4852 (1.3533) acc 96.8750 (96.4062) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4411 (1.3829) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 843.012, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6940 (1.3428) acc 87.5000 (96.8750) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.3722 (1.3733) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 842.895, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6812 (1.3573) acc 87.5000 (95.9375) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.3823 (1.3609) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 842.933, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6520 (1.3726) acc 93.7500 (96.0938) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:19 loss 1.4473 (1.3941) acc 87.5000 (94.9219) lr 0.260000 -FPS@all 842.931, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:12:18 loss 1.4425 (1.3580) acc 93.7500 (95.7812) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:12:18 loss 1.4277 (1.3738) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 843.119, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:19 loss 1.6428 (1.3625) acc 90.6250 (96.8750) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4026 (1.3752) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 843.043, TIME@all 0.304 -epoch: [65/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:12:18 loss 1.4454 (1.3404) acc 93.7500 (97.0312) lr 0.260000 -epoch: [65/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:12:18 loss 1.4990 (1.3608) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 843.277, TIME@all 0.304 -epoch: [66/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:12:09 loss 1.2669 (1.3142) acc 100.0000 (97.0312) lr 0.260000 -epoch: [66/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.3008 (1.3549) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 841.919, TIME@all 0.304 -epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:12:08 loss 1.4964 (1.2883) acc 87.5000 (97.8125) lr 0.260000 -epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.2119 (1.3021) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 841.978, TIME@all 0.304 -epoch: [66/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:12:09 loss 1.4019 (1.2943) acc 93.7500 (98.2812) lr 0.260000 -epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.2801 (1.3310) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 841.872, TIME@all 0.304 -epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.6525 (1.3076) acc 90.6250 (96.8750) lr 0.260000 -epoch: [66/350][40/50] time 0.308 (0.304) data 0.001 (0.006) eta 1:12:06 loss 1.3389 (1.3433) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 841.934, TIME@all 0.304 -epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.3781 (1.3221) acc 93.7500 (96.5625) lr 0.260000 -epoch: [66/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 1:12:05 loss 1.2421 (1.3417) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 842.101, TIME@all 0.304 -epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.2878 (1.2871) acc 100.0000 (98.2812) lr 0.260000 -epoch: [66/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.2744 (1.3262) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 841.916, TIME@all 0.304 -epoch: [66/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:12:08 loss 1.2692 (1.2847) acc 100.0000 (98.1250) lr 0.260000 -epoch: [66/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:12:06 loss 1.2470 (1.3130) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 842.238, TIME@all 0.304 -epoch: [66/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:12:08 loss 1.4780 (1.3149) acc 96.8750 (97.8125) lr 0.260000 -epoch: [66/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 1:12:05 loss 1.3510 (1.3300) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.043, TIME@all 0.304 -epoch: [67/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 1:11:56 loss 1.6315 (1.3038) acc 90.6250 (97.8125) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.4039 (1.3411) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 842.706, TIME@all 0.304 -epoch: [67/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:12:00 loss 1.4870 (1.3153) acc 96.8750 (97.5000) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.3249 (1.3300) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 842.750, TIME@all 0.304 -epoch: [67/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 1:11:57 loss 1.5381 (1.3011) acc 87.5000 (97.0312) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:11:47 loss 1.3782 (1.3322) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 842.747, TIME@all 0.304 -epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:57 loss 1.3772 (1.2872) acc 90.6250 (96.5625) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3688 (1.3271) acc 87.5000 (95.9375) lr 0.260000 -FPS@all 842.713, TIME@all 0.304 -epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:57 loss 1.4402 (1.2910) acc 93.7500 (97.8125) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3249 (1.3203) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 842.726, TIME@all 0.304 -epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.5720 (1.3029) acc 87.5000 (96.7188) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:46 loss 1.4102 (1.3204) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 842.913, TIME@all 0.304 -epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.4870 (1.3240) acc 90.6250 (95.6250) lr 0.260000 -epoch: [67/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3048 (1.3424) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 843.057, TIME@all 0.304 -epoch: [67/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 1:11:56 loss 1.4612 (1.3030) acc 90.6250 (96.8750) lr 0.260000 -epoch: [67/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:11:47 loss 1.3369 (1.3237) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 842.842, TIME@all 0.304 -epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.2848 (1.2666) acc 96.8750 (97.9688) lr 0.260000 -epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.4895 (1.3024) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 842.769, TIME@all 0.304 -epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.3383 (1.2770) acc 96.8750 (98.4375) lr 0.260000 -epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.2673 (1.3096) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 842.800, TIME@all 0.304 -epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:37 loss 1.2791 (1.2641) acc 96.8750 (98.2812) lr 0.260000 -epoch: [68/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 1:11:33 loss 1.2372 (1.2824) acc 100.0000 (97.9688) lr 0.260000 -FPS@all 842.718, TIME@all 0.304 -epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:37 loss 1.2076 (1.2802) acc 100.0000 (98.4375) lr 0.260000 -epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.2608 (1.3206) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.767, TIME@all 0.304 -epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:36 loss 1.4035 (1.2842) acc 96.8750 (97.8125) lr 0.260000 -epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:32 loss 1.2223 (1.3043) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 842.940, TIME@all 0.304 -epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:36 loss 1.1997 (1.2662) acc 100.0000 (97.9688) lr 0.260000 -epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:32 loss 1.3210 (1.3103) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.899, TIME@all 0.304 -epoch: [68/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:11:37 loss 1.3303 (1.2908) acc 96.8750 (97.9688) lr 0.260000 -epoch: [68/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.4232 (1.3196) acc 90.6250 (97.4219) lr 0.260000 -FPS@all 842.755, TIME@all 0.304 -epoch: [68/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:39 loss 1.2603 (1.2626) acc 100.0000 (97.6562) lr 0.260000 -epoch: [68/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:11:33 loss 1.5740 (1.3183) acc 87.5000 (96.5625) lr 0.260000 -FPS@all 842.892, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2599 (1.2779) acc 100.0000 (98.4375) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3795 (1.3087) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.671, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.4603 (1.2825) acc 93.7500 (97.3438) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3696 (1.3033) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.617, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2314 (1.2696) acc 100.0000 (98.9062) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.2859 (1.3046) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 842.718, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.3061 (1.2763) acc 96.8750 (97.8125) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.3492 (1.3122) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.643, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:15 loss 1.2660 (1.2790) acc 100.0000 (98.1250) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.4566 (1.2915) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 842.670, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:14 loss 1.2204 (1.2810) acc 100.0000 (98.2812) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 1:11:21 loss 1.3764 (1.3303) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.827, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:11:14 loss 1.3620 (1.2916) acc 96.8750 (97.5000) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 1:11:21 loss 1.4330 (1.3189) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 842.856, TIME@all 0.304 -epoch: [69/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:11:14 loss 1.5021 (1.3063) acc 93.7500 (97.0312) lr 0.260000 -epoch: [69/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 1:11:22 loss 1.2842 (1.3214) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 843.068, TIME@all 0.304 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.6224 (1.2752) acc 87.5000 (97.9688) lr 0.260000 -epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:29 loss 1.2994 (1.3150) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 839.053, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.3359 (1.2756) acc 93.7500 (97.8125) lr 0.260000 -epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:29 loss 1.2923 (1.3133) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 839.021, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.012) eta 1:11:39 loss 1.3040 (1.2631) acc 93.7500 (98.7500) lr 0.260000 -epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 1:11:28 loss 1.4032 (1.3087) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 839.090, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:39 loss 1.4123 (1.3330) acc 93.7500 (97.0312) lr 0.260000 -epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.2989 (1.3425) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 839.037, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.014) eta 1:11:38 loss 1.3629 (1.2684) acc 96.8750 (98.1250) lr 0.260000 -epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:27 loss 1.3690 (1.2976) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 839.239, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:39 loss 1.5699 (1.2876) acc 90.6250 (97.3438) lr 0.260000 -epoch: [70/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 1:11:29 loss 1.3177 (1.3113) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 839.049, TIME@all 0.305 -epoch: [70/350][20/50] time 0.308 (0.306) data 0.000 (0.013) eta 1:11:38 loss 1.4117 (1.2602) acc 93.7500 (98.1250) lr 0.260000 -epoch: [70/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.3073 (1.2949) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 839.175, TIME@all 0.305 -epoch: [70/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 1:11:40 loss 1.2623 (1.2744) acc 96.8750 (98.4375) lr 0.260000 -epoch: [70/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 1:11:28 loss 1.5246 (1.3314) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 839.388, TIME@all 0.305 -epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.011) eta 1:10:57 loss 1.4522 (1.4024) acc 90.6250 (94.6875) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.3316 (1.4615) acc 100.0000 (93.9062) lr 0.260000 -FPS@all 843.159, TIME@all 0.304 -epoch: [71/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.6315 (1.3950) acc 90.6250 (94.6875) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 1:10:46 loss 1.6204 (1.4642) acc 90.6250 (93.3594) lr 0.260000 -FPS@all 843.216, TIME@all 0.304 -epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5799 (1.4491) acc 90.6250 (94.0625) lr 0.260000 -epoch: [71/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.2984 (1.4894) acc 100.0000 (94.2188) lr 0.260000 -FPS@all 843.226, TIME@all 0.304 -epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:58 loss 1.4175 (1.3849) acc 96.8750 (95.6250) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.6282 (1.4559) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 843.190, TIME@all 0.304 -epoch: [71/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:10:56 loss 1.4903 (1.3910) acc 96.8750 (95.6250) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 1:10:45 loss 1.3971 (1.4619) acc 93.7500 (93.7500) lr 0.260000 -FPS@all 843.394, TIME@all 0.304 -epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:10:56 loss 1.5594 (1.4099) acc 90.6250 (94.8438) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:45 loss 1.4386 (1.4697) acc 93.7500 (93.6719) lr 0.260000 -FPS@all 843.358, TIME@all 0.304 -epoch: [71/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5365 (1.4119) acc 90.6250 (95.0000) lr 0.260000 -epoch: [71/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.4813 (1.4704) acc 87.5000 (93.2812) lr 0.260000 -FPS@all 843.189, TIME@all 0.304 -epoch: [71/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:10:57 loss 1.5959 (1.3871) acc 90.6250 (94.8438) lr 0.260000 -epoch: [71/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 1:10:46 loss 1.5447 (1.4605) acc 87.5000 (93.5938) lr 0.260000 -FPS@all 843.544, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:10:36 loss 1.4566 (1.3945) acc 93.7500 (95.6250) lr 0.260000 -epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:10:25 loss 1.3345 (1.4249) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 843.810, TIME@all 0.303 -epoch: [72/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:10:36 loss 1.4857 (1.3731) acc 90.6250 (95.1562) lr 0.260000 -epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:10:25 loss 1.3887 (1.4192) acc 100.0000 (94.6094) lr 0.260000 -FPS@all 843.858, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:10:35 loss 1.5647 (1.4009) acc 90.6250 (94.8438) lr 0.260000 -epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:10:24 loss 1.3360 (1.4226) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 843.898, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:35 loss 1.3431 (1.3804) acc 96.8750 (96.4062) lr 0.260000 -epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:10:24 loss 1.5825 (1.4075) acc 90.6250 (95.5469) lr 0.260000 -FPS@all 843.936, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:35 loss 1.3913 (1.3710) acc 96.8750 (95.7812) lr 0.260000 -epoch: [72/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.3815 (1.4063) acc 96.8750 (95.0781) lr 0.260000 -FPS@all 844.207, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:10:35 loss 1.5704 (1.3970) acc 93.7500 (94.3750) lr 0.260000 -epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:10:24 loss 1.3494 (1.4203) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 844.029, TIME@all 0.303 -epoch: [72/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:10:36 loss 1.3494 (1.3796) acc 96.8750 (95.7812) lr 0.260000 -epoch: [72/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.5279 (1.4292) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 843.822, TIME@all 0.303 -epoch: [72/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:10:36 loss 1.4673 (1.4009) acc 93.7500 (95.7812) lr 0.260000 -epoch: [72/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:10:25 loss 1.4637 (1.4466) acc 100.0000 (94.4531) lr 0.260000 -FPS@all 843.844, TIME@all 0.303 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5737 (1.3798) acc 90.6250 (95.0000) lr 0.260000 -epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.4560 (1.3907) acc 93.7500 (94.7656) lr 0.260000 -FPS@all 843.065, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:24 loss 1.6931 (1.3467) acc 93.7500 (96.2500) lr 0.260000 -epoch: [73/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.2662 (1.3898) acc 100.0000 (95.4688) lr 0.260000 -FPS@all 843.124, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:22 loss 1.3951 (1.3423) acc 93.7500 (96.8750) lr 0.260000 -epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:10:15 loss 1.3576 (1.3808) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 843.302, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:24 loss 1.4726 (1.3285) acc 93.7500 (97.0312) lr 0.260000 -epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.3291 (1.3577) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 843.072, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5604 (1.3383) acc 87.5000 (96.2500) lr 0.260000 -epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.006) eta 1:10:16 loss 1.3358 (1.3643) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 843.134, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:10:22 loss 1.9414 (1.3586) acc 81.2500 (96.5625) lr 0.260000 -epoch: [73/350][40/50] time 0.297 (0.304) data 0.000 (0.007) eta 1:10:16 loss 1.3189 (1.3808) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 843.216, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.4142 (1.3190) acc 96.8750 (97.3438) lr 0.260000 -epoch: [73/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:10:16 loss 1.3890 (1.3660) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 843.361, TIME@all 0.304 -epoch: [73/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:10:23 loss 1.5093 (1.3436) acc 87.5000 (97.0312) lr 0.260000 -epoch: [73/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 1:10:17 loss 1.3533 (1.3593) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 843.103, TIME@all 0.304 -epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:09:57 loss 1.5269 (1.3388) acc 87.5000 (96.8750) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.4942 (1.3514) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 843.468, TIME@all 0.304 -epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:09:58 loss 1.3986 (1.3229) acc 93.7500 (96.5625) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.4687 (1.3485) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 843.495, TIME@all 0.303 -epoch: [74/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 1:09:57 loss 1.4649 (1.3232) acc 93.7500 (95.9375) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:09:56 loss 1.6049 (1.3505) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 843.536, TIME@all 0.303 -epoch: [74/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 1:09:56 loss 1.8031 (1.3395) acc 87.5000 (96.7188) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:55 loss 1.5689 (1.3476) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 843.680, TIME@all 0.303 -epoch: [74/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:09:57 loss 1.3483 (1.3169) acc 96.8750 (97.0312) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.8912 (1.3613) acc 81.2500 (95.7812) lr 0.260000 -FPS@all 843.803, TIME@all 0.303 -epoch: [74/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:09:57 loss 1.6669 (1.3559) acc 90.6250 (97.0312) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.5122 (1.3670) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 843.450, TIME@all 0.304 -epoch: [74/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:09:58 loss 1.3065 (1.3187) acc 100.0000 (97.8125) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:09:56 loss 1.7678 (1.3600) acc 81.2500 (96.5625) lr 0.260000 -FPS@all 843.446, TIME@all 0.304 -epoch: [74/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 1:09:57 loss 1.3488 (1.3139) acc 96.8750 (97.3438) lr 0.260000 -epoch: [74/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:09:56 loss 1.4525 (1.3542) acc 90.6250 (95.5469) lr 0.260000 -FPS@all 843.625, TIME@all 0.303 -epoch: [75/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:09:54 loss 1.6531 (1.3441) acc 90.6250 (96.8750) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.3539 (1.3823) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 843.044, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:53 loss 1.5347 (1.3426) acc 90.6250 (95.6250) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.4794 (1.3756) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 843.103, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:53 loss 1.3439 (1.3282) acc 96.8750 (96.8750) lr 0.260000 -epoch: [75/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:09:44 loss 1.5987 (1.3789) acc 87.5000 (95.3125) lr 0.260000 -FPS@all 843.109, TIME@all 0.304 -epoch: [75/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:09:54 loss 1.4990 (1.3361) acc 96.8750 (97.3438) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:45 loss 1.4720 (1.3733) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 843.052, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 1:09:53 loss 1.4803 (1.3099) acc 90.6250 (97.8125) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:44 loss 1.4857 (1.3691) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 843.236, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:54 loss 1.3383 (1.3448) acc 100.0000 (96.7188) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:45 loss 1.3479 (1.3667) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 843.034, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:09:53 loss 1.4098 (1.3335) acc 87.5000 (97.0312) lr 0.260000 -epoch: [75/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:44 loss 1.3136 (1.3807) acc 96.8750 (95.3906) lr 0.260000 -FPS@all 843.184, TIME@all 0.304 -epoch: [75/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:09:52 loss 1.4255 (1.3198) acc 93.7500 (96.7188) lr 0.260000 -epoch: [75/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 1:09:44 loss 1.4892 (1.3904) acc 90.6250 (95.0000) lr 0.260000 -FPS@all 843.400, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:40 loss 1.2829 (1.2731) acc 100.0000 (99.0625) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:28 loss 1.3157 (1.3007) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.454, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:39 loss 1.3511 (1.2733) acc 93.7500 (98.4375) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.3461 (1.3067) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 843.469, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:09:39 loss 1.2740 (1.2828) acc 96.8750 (96.4062) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:09:28 loss 1.2339 (1.3072) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 843.395, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:40 loss 1.2665 (1.2598) acc 100.0000 (99.0625) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:28 loss 1.3039 (1.3007) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.434, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:09:39 loss 1.2532 (1.2859) acc 100.0000 (98.2812) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:09:28 loss 1.2610 (1.3036) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 843.438, TIME@all 0.304 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:09:38 loss 1.2981 (1.2839) acc 93.7500 (97.5000) lr 0.260000 -epoch: [76/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:09:27 loss 1.2194 (1.3139) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 843.635, TIME@all 0.303 -epoch: [76/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:09:39 loss 1.2465 (1.2814) acc 96.8750 (98.1250) lr 0.260000 -epoch: [76/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.2333 (1.3081) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 843.558, TIME@all 0.303 -epoch: [76/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:09:38 loss 1.3763 (1.3055) acc 93.7500 (97.5000) lr 0.260000 -epoch: [76/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:09:27 loss 1.3706 (1.3125) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 843.800, TIME@all 0.303 -epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.3218 (1.2782) acc 93.7500 (97.3438) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.3280 (1.3258) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 843.916, TIME@all 0.303 -epoch: [77/350][20/50] time 0.300 (0.304) data 0.001 (0.014) eta 1:09:20 loss 1.3053 (1.2936) acc 100.0000 (97.1875) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4440 (1.3263) acc 90.6250 (96.1719) lr 0.260000 -FPS@all 843.960, TIME@all 0.303 -epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.2046 (1.2731) acc 100.0000 (98.7500) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4825 (1.3150) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 844.005, TIME@all 0.303 -epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:09:20 loss 1.2135 (1.2899) acc 96.8750 (96.2500) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.4221 (1.3293) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 843.946, TIME@all 0.303 -epoch: [77/350][20/50] time 0.299 (0.304) data 0.001 (0.014) eta 1:09:18 loss 1.3116 (1.2877) acc 96.8750 (98.1250) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:10 loss 1.3180 (1.3177) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 844.148, TIME@all 0.303 -epoch: [77/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 1:09:20 loss 1.2814 (1.2954) acc 96.8750 (98.1250) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.2931 (1.3301) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.931, TIME@all 0.303 -epoch: [77/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 1:09:19 loss 1.1955 (1.2923) acc 100.0000 (97.8125) lr 0.260000 -epoch: [77/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 1:09:11 loss 1.2827 (1.3163) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 844.090, TIME@all 0.303 -epoch: [77/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 1:09:21 loss 1.2381 (1.2930) acc 100.0000 (97.3438) lr 0.260000 -epoch: [77/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:09:11 loss 1.3928 (1.3168) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 844.249, TIME@all 0.303 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:03 loss 1.3823 (1.2845) acc 93.7500 (96.7188) lr 0.260000 -epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:02 loss 1.3412 (1.3144) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 841.853, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:03 loss 1.3149 (1.2799) acc 93.7500 (97.6562) lr 0.260000 -epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2757 (1.3155) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 841.808, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:09:04 loss 1.3501 (1.3007) acc 90.6250 (97.8125) lr 0.260000 -epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2856 (1.3213) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 841.796, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:09:02 loss 1.3553 (1.2879) acc 96.8750 (97.5000) lr 0.260000 -epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:09:02 loss 1.4267 (1.3170) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 842.024, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.3148 (1.2841) acc 93.7500 (97.9688) lr 0.260000 -epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.3230 (1.3068) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 841.849, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.2928 (1.2980) acc 96.8750 (97.3438) lr 0.260000 -epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.2648 (1.3272) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 841.823, TIME@all 0.304 -epoch: [78/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:09:03 loss 1.3071 (1.2895) acc 96.8750 (97.5000) lr 0.260000 -epoch: [78/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:09:02 loss 1.4421 (1.3341) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 841.954, TIME@all 0.304 -epoch: [78/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:09:02 loss 1.3070 (1.2848) acc 93.7500 (97.3438) lr 0.260000 -epoch: [78/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:09:03 loss 1.3565 (1.3178) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 842.176, TIME@all 0.304 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 1:08:47 loss 1.3355 (1.2888) acc 96.8750 (97.9688) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.4211 (1.3134) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 843.440, TIME@all 0.304 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:47 loss 1.2400 (1.2884) acc 96.8750 (97.1875) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.2535 (1.3106) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 843.435, TIME@all 0.304 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:46 loss 1.3063 (1.2949) acc 96.8750 (97.0312) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:41 loss 1.3666 (1.3190) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 843.506, TIME@all 0.303 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:08:47 loss 1.5164 (1.3154) acc 96.8750 (97.0312) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:08:42 loss 1.4469 (1.3136) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 843.422, TIME@all 0.304 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:08:47 loss 1.4266 (1.3152) acc 93.7500 (96.5625) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:42 loss 1.3240 (1.3268) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 843.427, TIME@all 0.304 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:08:46 loss 1.3906 (1.2976) acc 96.8750 (98.1250) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:40 loss 1.2904 (1.3190) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 843.622, TIME@all 0.303 -epoch: [79/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 1:08:46 loss 1.4550 (1.2889) acc 90.6250 (97.5000) lr 0.260000 -epoch: [79/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:41 loss 1.3224 (1.3035) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 843.575, TIME@all 0.303 -epoch: [79/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:08:46 loss 1.3765 (1.2776) acc 96.8750 (97.9688) lr 0.260000 -epoch: [79/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:08:41 loss 1.5199 (1.3067) acc 87.5000 (97.1875) lr 0.260000 -FPS@all 843.832, TIME@all 0.303 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.3883 (1.3154) acc 96.8750 (97.0312) lr 0.260000 -epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:38 loss 1.2750 (1.3586) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 841.271, TIME@all 0.304 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.2339 (1.2560) acc 100.0000 (98.7500) lr 0.260000 -epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:39 loss 1.3034 (1.3180) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 841.222, TIME@all 0.304 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:08:45 loss 1.2090 (1.3241) acc 100.0000 (96.8750) lr 0.260000 -epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:08:38 loss 1.3720 (1.3582) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 841.270, TIME@all 0.304 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.3533 (1.2988) acc 96.8750 (97.3438) lr 0.260000 -epoch: [80/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:08:39 loss 1.3261 (1.3431) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 841.258, TIME@all 0.304 -epoch: [80/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:08:46 loss 1.1980 (1.2917) acc 100.0000 (97.3438) lr 0.260000 -epoch: [80/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:08:39 loss 1.2387 (1.3384) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 841.239, TIME@all 0.304 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.1775 (1.2928) acc 100.0000 (98.1250) lr 0.260000 -epoch: [80/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 1:08:38 loss 1.2862 (1.3325) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 841.430, TIME@all 0.304 -epoch: [80/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:08:45 loss 1.3085 (1.2993) acc 96.8750 (97.6562) lr 0.260000 -epoch: [80/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 1:08:38 loss 1.3851 (1.3243) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 841.379, TIME@all 0.304 -epoch: [80/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:08:43 loss 1.4225 (1.3103) acc 96.8750 (96.8750) lr 0.260000 -epoch: [80/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:08:39 loss 1.2010 (1.3449) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 841.623, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:23 loss 1.3057 (1.2922) acc 96.8750 (97.5000) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2956 (1.3065) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 841.310, TIME@all 0.304 -epoch: [81/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.3001 (1.2955) acc 100.0000 (97.9688) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.3819 (1.3212) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 841.351, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.2767 (1.2857) acc 100.0000 (98.1250) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2678 (1.3229) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 841.396, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:23 loss 1.4151 (1.3208) acc 93.7500 (97.8125) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.3037 (1.3250) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 841.343, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:08:22 loss 1.2602 (1.3154) acc 100.0000 (96.7188) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:08:21 loss 1.2552 (1.3436) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 841.357, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:08:21 loss 1.4395 (1.3001) acc 96.8750 (97.6562) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:08:20 loss 1.2505 (1.2963) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 841.541, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:08:22 loss 1.2372 (1.3116) acc 96.8750 (97.3438) lr 0.260000 -epoch: [81/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:08:20 loss 1.2938 (1.3323) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 841.484, TIME@all 0.304 -epoch: [81/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:08:24 loss 1.3377 (1.3121) acc 100.0000 (97.5000) lr 0.260000 -epoch: [81/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 1:08:22 loss 1.3792 (1.3189) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 841.501, TIME@all 0.304 -epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:08:13 loss 1.3166 (1.3182) acc 100.0000 (97.0312) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:07:59 loss 1.3478 (1.3602) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 842.646, TIME@all 0.304 -epoch: [82/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.5052 (1.3446) acc 96.8750 (96.5625) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.3614 (1.3573) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 842.623, TIME@all 0.304 -epoch: [82/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.4148 (1.3096) acc 87.5000 (97.1875) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.3269 (1.3328) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.710, TIME@all 0.304 -epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:08:13 loss 1.3395 (1.3186) acc 100.0000 (96.5625) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.3500 (1.3602) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 842.649, TIME@all 0.304 -epoch: [82/350][20/50] time 0.308 (0.305) data 0.001 (0.013) eta 1:08:14 loss 1.4189 (1.3051) acc 96.8750 (97.1875) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:08:00 loss 1.4011 (1.3526) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 842.855, TIME@all 0.304 -epoch: [82/350][20/50] time 0.307 (0.305) data 0.001 (0.014) eta 1:08:12 loss 1.4017 (1.3010) acc 93.7500 (97.3438) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.5075 (1.3643) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 842.859, TIME@all 0.304 -epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:08:13 loss 1.3687 (1.3137) acc 96.8750 (97.0312) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:59 loss 1.2253 (1.3500) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 842.663, TIME@all 0.304 -epoch: [82/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 1:08:13 loss 1.5252 (1.3092) acc 96.8750 (97.8125) lr 0.260000 -epoch: [82/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:58 loss 1.4037 (1.3336) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.766, TIME@all 0.304 -epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3078 (1.2577) acc 100.0000 (98.4375) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.3339 (1.3105) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.425, TIME@all 0.304 -epoch: [83/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:56 loss 1.2924 (1.2884) acc 96.8750 (98.1250) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:07:43 loss 1.1865 (1.3261) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 843.349, TIME@all 0.304 -epoch: [83/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:55 loss 1.2475 (1.2712) acc 100.0000 (98.4375) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:07:43 loss 1.3800 (1.3073) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 843.433, TIME@all 0.304 -epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:54 loss 1.3466 (1.2769) acc 93.7500 (97.9688) lr 0.260000 -epoch: [83/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:07:42 loss 1.2942 (1.3067) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 843.587, TIME@all 0.303 -epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.2896 (1.2674) acc 96.8750 (98.1250) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2962 (1.3212) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 843.384, TIME@all 0.304 -epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3808 (1.2604) acc 96.8750 (98.4375) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2489 (1.2942) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.387, TIME@all 0.304 -epoch: [83/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:07:55 loss 1.3395 (1.2778) acc 90.6250 (97.1875) lr 0.260000 -epoch: [83/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:07:42 loss 1.2456 (1.3142) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 843.519, TIME@all 0.303 -epoch: [83/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:07:54 loss 1.3370 (1.2709) acc 96.8750 (98.2812) lr 0.260000 -epoch: [83/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:43 loss 1.2786 (1.2993) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.667, TIME@all 0.303 -epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.2512 (1.2788) acc 100.0000 (97.8125) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:29 loss 1.4104 (1.3038) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.489, TIME@all 0.304 -epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.5222 (1.3292) acc 93.7500 (96.2500) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:28 loss 1.2064 (1.3408) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 842.538, TIME@all 0.304 -epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:07:33 loss 1.2877 (1.3002) acc 100.0000 (97.5000) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:07:29 loss 1.2356 (1.3220) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 842.476, TIME@all 0.304 -epoch: [84/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:07:33 loss 1.3719 (1.3026) acc 96.8750 (97.3438) lr 0.260000 -epoch: [84/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 1:07:29 loss 1.3359 (1.3285) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 842.514, TIME@all 0.304 -epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:07:32 loss 1.4282 (1.3030) acc 96.8750 (97.6562) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:28 loss 1.2933 (1.3273) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 842.702, TIME@all 0.304 -epoch: [84/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:07:33 loss 1.2960 (1.2757) acc 100.0000 (98.2812) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:07:29 loss 1.3701 (1.3068) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 842.499, TIME@all 0.304 -epoch: [84/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:07:32 loss 1.2809 (1.2786) acc 100.0000 (98.2812) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:07:28 loss 1.2738 (1.3098) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 842.631, TIME@all 0.304 -epoch: [84/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:07:31 loss 1.2653 (1.2851) acc 100.0000 (97.3438) lr 0.260000 -epoch: [84/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 1:07:28 loss 1.2460 (1.3083) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 842.890, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 1:07:24 loss 1.5043 (1.3019) acc 84.3750 (96.5625) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.4409 (1.3296) acc 87.5000 (96.0938) lr 0.260000 -FPS@all 842.096, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:07:24 loss 1.4813 (1.3283) acc 93.7500 (96.4062) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.3964 (1.3490) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 842.033, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 1:07:24 loss 1.5987 (1.3423) acc 87.5000 (96.4062) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.4993 (1.3449) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 842.128, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.3742 (1.3277) acc 90.6250 (97.1875) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:07:18 loss 1.3074 (1.3383) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 842.216, TIME@all 0.304 -epoch: [85/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.4078 (1.3041) acc 93.7500 (97.0312) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:07:18 loss 1.4362 (1.3440) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 842.058, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.6190 (1.3200) acc 90.6250 (97.0312) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:07:18 loss 1.3078 (1.3520) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 842.072, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 1:07:24 loss 1.5979 (1.3283) acc 90.6250 (96.2500) lr 0.260000 -epoch: [85/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:07:17 loss 1.4457 (1.3576) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 842.423, TIME@all 0.304 -epoch: [85/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 1:07:23 loss 1.5305 (1.3236) acc 87.5000 (96.4062) lr 0.260000 -epoch: [85/350][40/50] time 0.307 (0.305) data 0.001 (0.007) eta 1:07:17 loss 1.5103 (1.3593) acc 84.3750 (95.5469) lr 0.260000 -FPS@all 842.269, TIME@all 0.304 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.3680 (1.3006) acc 93.7500 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.4668 (1.3404) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 843.962, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:58 loss 1.2946 (1.3033) acc 96.8750 (97.0312) lr 0.260000 -epoch: [86/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3861 (1.3354) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 843.908, TIME@all 0.303 -epoch: [86/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 1:06:56 loss 1.5178 (1.3082) acc 93.7500 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:06:53 loss 1.2489 (1.3342) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 844.034, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.6624 (1.3236) acc 87.5000 (96.2500) lr 0.260000 -epoch: [86/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3353 (1.3395) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 843.975, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:06:56 loss 1.4601 (1.3011) acc 84.3750 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 1:06:52 loss 1.4721 (1.3435) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 844.144, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:57 loss 1.6571 (1.3200) acc 90.6250 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:53 loss 1.3458 (1.3413) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 843.964, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:56 loss 1.5813 (1.3161) acc 90.6250 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:06:52 loss 1.5208 (1.3536) acc 90.6250 (95.9375) lr 0.260000 -FPS@all 844.101, TIME@all 0.303 -epoch: [86/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:06:56 loss 1.5266 (1.3019) acc 93.7500 (97.5000) lr 0.260000 -epoch: [86/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:52 loss 1.5565 (1.3394) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 844.327, TIME@all 0.303 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:07:03 loss 1.2506 (1.2908) acc 100.0000 (97.9688) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.006) eta 1:06:53 loss 1.2087 (1.3390) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 841.067, TIME@all 0.304 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:03 loss 1.2664 (1.3017) acc 100.0000 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.3215 (1.3446) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 841.082, TIME@all 0.304 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.3143 (1.3196) acc 96.8750 (96.7188) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2395 (1.3516) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 841.096, TIME@all 0.304 -epoch: [87/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2574 (1.3028) acc 96.8750 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.3102 (1.3588) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 841.075, TIME@all 0.304 -epoch: [87/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 1:07:02 loss 1.2778 (1.3134) acc 96.8750 (97.9688) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.3752 (1.3627) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 841.222, TIME@all 0.304 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2818 (1.3144) acc 96.8750 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:53 loss 1.2411 (1.3468) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 841.084, TIME@all 0.304 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:01 loss 1.4030 (1.3384) acc 93.7500 (96.2500) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2569 (1.3605) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 841.271, TIME@all 0.304 -epoch: [87/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:07:02 loss 1.2732 (1.2884) acc 100.0000 (97.8125) lr 0.260000 -epoch: [87/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:06:52 loss 1.2440 (1.3239) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 841.492, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.4015 (1.3529) acc 93.7500 (95.7812) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.5114 (1.3771) acc 96.8750 (95.0000) lr 0.260000 -FPS@all 842.459, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:39 loss 1.5629 (1.3676) acc 90.6250 (96.2500) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:30 loss 1.4610 (1.3882) acc 90.6250 (95.6250) lr 0.260000 -FPS@all 842.455, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.001 (0.013) eta 1:06:38 loss 1.3680 (1.3105) acc 93.7500 (97.6562) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:28 loss 1.3638 (1.3553) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 842.674, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:39 loss 1.2311 (1.3024) acc 96.8750 (96.7188) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:29 loss 1.3175 (1.3453) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 842.450, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 1:06:39 loss 1.3159 (1.3216) acc 96.8750 (96.8750) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.5484 (1.3424) acc 90.6250 (96.0938) lr 0.260000 -FPS@all 842.485, TIME@all 0.304 -epoch: [88/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 1:06:38 loss 1.4833 (1.3210) acc 87.5000 (96.8750) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:29 loss 1.3982 (1.3515) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 842.614, TIME@all 0.304 -epoch: [88/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.3303 (1.2922) acc 96.8750 (98.2812) lr 0.260000 -epoch: [88/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.3706 (1.3449) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 842.516, TIME@all 0.304 -epoch: [88/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:06:38 loss 1.3866 (1.2928) acc 93.7500 (96.8750) lr 0.260000 -epoch: [88/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:06:29 loss 1.4547 (1.3428) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 842.826, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 1:06:21 loss 1.3240 (1.2887) acc 93.7500 (97.8125) lr 0.260000 -epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.4713 (1.3166) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 842.486, TIME@all 0.304 -epoch: [89/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.2727 (1.2728) acc 100.0000 (98.5938) lr 0.260000 -epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.3718 (1.3193) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 842.559, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.3062 (1.2529) acc 96.8750 (99.2188) lr 0.260000 -epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.4767 (1.2991) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 842.507, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.3359 (1.2681) acc 96.8750 (97.6562) lr 0.260000 -epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.3399 (1.3054) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 842.499, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:06:20 loss 1.2950 (1.2754) acc 93.7500 (98.2812) lr 0.260000 -epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:06:12 loss 1.3956 (1.3040) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 842.717, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:21 loss 1.6442 (1.2824) acc 87.5000 (98.4375) lr 0.260000 -epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:13 loss 1.6008 (1.3052) acc 90.6250 (97.6562) lr 0.260000 -FPS@all 842.522, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.3531 (1.2718) acc 100.0000 (98.4375) lr 0.260000 -epoch: [89/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:06:12 loss 1.3198 (1.3012) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 842.668, TIME@all 0.304 -epoch: [89/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:06:20 loss 1.1990 (1.2923) acc 100.0000 (97.8125) lr 0.260000 -epoch: [89/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:06:12 loss 1.4604 (1.3117) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 842.934, TIME@all 0.304 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:15 loss 1.2458 (1.3240) acc 96.8750 (97.5000) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.4084 (1.3597) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 840.598, TIME@all 0.305 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:15 loss 1.4187 (1.3121) acc 93.7500 (97.0312) lr 0.260000 -epoch: [90/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3094 (1.3460) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 840.623, TIME@all 0.305 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 1:06:14 loss 1.3199 (1.3259) acc 96.8750 (96.4062) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3052 (1.3503) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 840.682, TIME@all 0.305 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 1:06:13 loss 1.2030 (1.3251) acc 100.0000 (97.1875) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:06:10 loss 1.2912 (1.3439) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 840.837, TIME@all 0.304 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.3056 (1.3133) acc 93.7500 (96.5625) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:11 loss 1.3316 (1.3708) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 840.584, TIME@all 0.305 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.3841 (1.3494) acc 96.8750 (95.9375) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:10 loss 1.2966 (1.3570) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 840.786, TIME@all 0.304 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:14 loss 1.2325 (1.3545) acc 100.0000 (96.2500) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.001 (0.006) eta 1:06:11 loss 1.2097 (1.3567) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 840.635, TIME@all 0.305 -epoch: [90/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 1:06:13 loss 1.3398 (1.3246) acc 96.8750 (96.8750) lr 0.260000 -epoch: [90/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:06:10 loss 1.2117 (1.3432) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 841.019, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.4597 (1.2901) acc 93.7500 (97.5000) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.5522 (1.3150) acc 87.5000 (96.7969) lr 0.260000 -FPS@all 842.822, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:50 loss 1.3264 (1.3039) acc 96.8750 (97.9688) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:42 loss 1.5275 (1.3374) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 842.977, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:51 loss 1.4144 (1.2841) acc 96.8750 (97.6562) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:05:43 loss 1.4038 (1.3115) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 842.840, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.3591 (1.2900) acc 96.8750 (97.5000) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.4482 (1.3142) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.857, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.5637 (1.3012) acc 93.7500 (97.1875) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3963 (1.3089) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 842.886, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:05:50 loss 1.5108 (1.3037) acc 84.3750 (97.0312) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:05:42 loss 1.3179 (1.3106) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 843.016, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:51 loss 1.3534 (1.2847) acc 96.8750 (98.2812) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3823 (1.2977) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.819, TIME@all 0.304 -epoch: [91/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:05:50 loss 1.3278 (1.2711) acc 96.8750 (97.5000) lr 0.260000 -epoch: [91/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:05:43 loss 1.3288 (1.3014) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 843.178, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:44 loss 1.2489 (1.2452) acc 100.0000 (98.4375) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4842 (1.3022) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 841.514, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.3949 (1.2695) acc 93.7500 (97.9688) lr 0.260000 -epoch: [92/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:05:34 loss 1.4699 (1.3021) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 841.573, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.3386 (1.2521) acc 93.7500 (97.9688) lr 0.260000 -epoch: [92/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4060 (1.2868) acc 93.7500 (97.9688) lr 0.260000 -FPS@all 841.584, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:44 loss 1.2568 (1.2437) acc 96.8750 (98.4375) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:33 loss 1.2903 (1.2765) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 841.528, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:43 loss 1.2163 (1.2608) acc 100.0000 (97.8125) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:32 loss 1.2906 (1.2825) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 841.721, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:44 loss 1.3547 (1.2651) acc 96.8750 (98.1250) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 1:05:33 loss 1.4468 (1.2912) acc 90.6250 (97.5781) lr 0.260000 -FPS@all 841.537, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 1:05:43 loss 1.2545 (1.2508) acc 100.0000 (97.9688) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.001 (0.006) eta 1:05:31 loss 1.3204 (1.2920) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.029, TIME@all 0.304 -epoch: [92/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 1:05:43 loss 1.3453 (1.2549) acc 93.7500 (98.2812) lr 0.260000 -epoch: [92/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 1:05:32 loss 1.2809 (1.2774) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 841.685, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:21 loss 1.4016 (1.2908) acc 90.6250 (97.3438) lr 0.260000 -epoch: [93/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3133 (1.2927) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 842.492, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:20 loss 1.3386 (1.3010) acc 96.8750 (97.5000) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3373 (1.3061) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.529, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:05:21 loss 1.2309 (1.2839) acc 96.8750 (97.8125) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:05:13 loss 1.3185 (1.3058) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 842.415, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:20 loss 1.2818 (1.2688) acc 96.8750 (97.8125) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:12 loss 1.4917 (1.3029) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 842.633, TIME@all 0.304 -epoch: [93/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.2983 (1.2861) acc 93.7500 (97.6562) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:13 loss 1.3063 (1.3116) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.464, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:20 loss 1.2772 (1.3016) acc 100.0000 (97.0312) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:12 loss 1.1988 (1.3352) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 842.603, TIME@all 0.304 -epoch: [93/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.4949 (1.3012) acc 93.7500 (97.6562) lr 0.260000 -epoch: [93/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:05:13 loss 1.3461 (1.3071) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.804, TIME@all 0.304 -epoch: [93/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:05:21 loss 1.2957 (1.3020) acc 96.8750 (96.2500) lr 0.260000 -epoch: [93/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:05:13 loss 1.2735 (1.3346) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 842.451, TIME@all 0.304 -epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 1:05:05 loss 1.2798 (1.2789) acc 93.7500 (97.9688) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:05:00 loss 1.2124 (1.3031) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 841.874, TIME@all 0.304 -epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.2559 (1.2881) acc 100.0000 (97.6562) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.3963 (1.3114) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 841.880, TIME@all 0.304 -epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:05:05 loss 1.3142 (1.2739) acc 96.8750 (98.5938) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:05:00 loss 1.2559 (1.3048) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 841.936, TIME@all 0.304 -epoch: [94/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3647 (1.2883) acc 93.7500 (96.7188) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 1:05:00 loss 1.2402 (1.3037) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 841.886, TIME@all 0.304 -epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3048 (1.2751) acc 100.0000 (97.6562) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.4740 (1.2982) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 842.196, TIME@all 0.304 -epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 1:05:04 loss 1.2275 (1.2785) acc 100.0000 (98.1250) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:04:59 loss 1.2926 (1.3063) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 842.081, TIME@all 0.304 -epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.3764 (1.2953) acc 93.7500 (97.0312) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:05:00 loss 1.2897 (1.2976) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 841.894, TIME@all 0.304 -epoch: [94/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:05:05 loss 1.2605 (1.2583) acc 100.0000 (97.9688) lr 0.260000 -epoch: [94/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:04:59 loss 1.4798 (1.3124) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 842.029, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2954 (1.2698) acc 96.8750 (97.9688) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 1:04:38 loss 1.3340 (1.3028) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 843.215, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2468 (1.2736) acc 100.0000 (98.7500) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:04:38 loss 1.3990 (1.3119) acc 93.7500 (97.8125) lr 0.260000 -FPS@all 843.181, TIME@all 0.304 -epoch: [95/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:44 loss 1.2942 (1.2780) acc 100.0000 (97.6562) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:04:38 loss 1.3878 (1.3065) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 843.266, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.2531 (1.2728) acc 100.0000 (97.9688) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.3919 (1.3215) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.215, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:43 loss 1.2275 (1.2497) acc 100.0000 (98.7500) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:37 loss 1.2781 (1.2791) acc 100.0000 (97.9688) lr 0.260000 -FPS@all 843.389, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.2736 (1.2613) acc 96.8750 (98.5938) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.2711 (1.2836) acc 96.8750 (98.2031) lr 0.260000 -FPS@all 843.219, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:43 loss 1.3501 (1.2765) acc 96.8750 (98.4375) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:37 loss 1.3989 (1.3184) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 843.350, TIME@all 0.304 -epoch: [95/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 1:04:44 loss 1.4287 (1.3046) acc 93.7500 (97.1875) lr 0.260000 -epoch: [95/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:04:38 loss 1.4178 (1.3230) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.638, TIME@all 0.303 -epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3958 (1.2964) acc 87.5000 (97.1875) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2558 (1.3348) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 842.938, TIME@all 0.304 -epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3680 (1.2969) acc 96.8750 (97.0312) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2060 (1.3043) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 842.983, TIME@all 0.304 -epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.4807 (1.2716) acc 93.7500 (97.3438) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.2345 (1.3121) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 843.009, TIME@all 0.304 -epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 1:04:29 loss 1.2844 (1.3103) acc 100.0000 (97.1875) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:04:25 loss 1.1798 (1.3054) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 843.140, TIME@all 0.304 -epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.4624 (1.2954) acc 90.6250 (97.5000) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.3258 (1.3243) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 842.953, TIME@all 0.304 -epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:04:30 loss 1.4190 (1.3087) acc 96.8750 (97.6562) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:04:26 loss 1.2931 (1.3126) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.970, TIME@all 0.304 -epoch: [96/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 1:04:30 loss 1.3106 (1.2681) acc 100.0000 (97.5000) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:04:26 loss 1.3530 (1.3053) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.241, TIME@all 0.304 -epoch: [96/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:04:29 loss 1.3375 (1.2693) acc 96.8750 (97.8125) lr 0.260000 -epoch: [96/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:04:25 loss 1.3358 (1.2889) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 843.114, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:04:30 loss 1.1695 (1.2635) acc 100.0000 (97.6562) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:04:21 loss 1.5265 (1.2895) acc 90.6250 (97.5000) lr 0.260000 -FPS@all 841.408, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:30 loss 1.2292 (1.2667) acc 96.8750 (98.4375) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.4445 (1.2901) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 841.432, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2617 (1.2498) acc 96.8750 (98.4375) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.2865 (1.2721) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 841.591, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:04:29 loss 1.2315 (1.2756) acc 100.0000 (97.6562) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 1:04:20 loss 1.3171 (1.2782) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 841.105, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2618 (1.2493) acc 100.0000 (98.9062) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.3217 (1.2797) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 841.551, TIME@all 0.304 -epoch: [97/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 1:04:29 loss 1.2591 (1.2625) acc 96.8750 (97.9688) lr 0.260000 -epoch: [97/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.4277 (1.3064) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 841.399, TIME@all 0.304 -epoch: [97/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 1:04:29 loss 1.1903 (1.2594) acc 100.0000 (97.8125) lr 0.260000 -epoch: [97/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 1:04:20 loss 1.3388 (1.2895) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 841.182, TIME@all 0.304 -epoch: [97/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 1:04:28 loss 1.2047 (1.2609) acc 100.0000 (98.1250) lr 0.260000 -epoch: [97/350][40/50] time 0.309 (0.305) data 0.000 (0.007) eta 1:04:20 loss 1.2923 (1.2992) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 841.670, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 1:03:55 loss 1.4851 (1.2749) acc 90.6250 (97.3438) lr 0.260000 -epoch: [98/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:58 loss 1.2640 (1.2886) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 842.343, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.3684 (1.2703) acc 93.7500 (98.2812) lr 0.260000 -epoch: [98/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:57 loss 1.4804 (1.2904) acc 90.6250 (97.5781) lr 0.260000 -FPS@all 842.362, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.4624 (1.2662) acc 93.7500 (98.1250) lr 0.260000 -epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.4303 (1.2901) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.409, TIME@all 0.304 -epoch: [98/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 1:03:55 loss 1.4632 (1.2617) acc 90.6250 (98.5938) lr 0.260000 -epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:58 loss 1.3854 (1.2998) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.339, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:55 loss 1.3994 (1.2891) acc 93.7500 (97.6562) lr 0.260000 -epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:58 loss 1.5809 (1.3074) acc 87.5000 (96.9531) lr 0.260000 -FPS@all 842.336, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 1:03:54 loss 1.3907 (1.2816) acc 93.7500 (97.0312) lr 0.260000 -epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.4892 (1.3030) acc 87.5000 (96.6406) lr 0.260000 -FPS@all 842.519, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.6576 (1.2833) acc 87.5000 (97.6562) lr 0.260000 -epoch: [98/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.2819 (1.2941) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 842.473, TIME@all 0.304 -epoch: [98/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 1:03:54 loss 1.4326 (1.2737) acc 93.7500 (97.6562) lr 0.260000 -epoch: [98/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:03:57 loss 1.5080 (1.2841) acc 90.6250 (97.5000) lr 0.260000 -FPS@all 842.709, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:54 loss 1.4885 (1.2922) acc 93.7500 (97.5000) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.3960 (1.3157) acc 90.6250 (96.9531) lr 0.260000 -FPS@all 842.556, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.3970 (1.2783) acc 96.8750 (97.3438) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2683 (1.3051) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 842.617, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.6742 (1.3117) acc 87.5000 (97.0312) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2714 (1.3211) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 842.532, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:53 loss 1.3677 (1.2969) acc 96.8750 (97.6562) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2363 (1.3082) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 842.552, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:52 loss 1.1993 (1.2978) acc 100.0000 (97.5000) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:41 loss 1.2642 (1.3220) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.756, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:03:53 loss 1.3145 (1.2703) acc 93.7500 (97.9688) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.3393 (1.3030) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.552, TIME@all 0.304 -epoch: [99/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 1:03:53 loss 1.4020 (1.2863) acc 93.7500 (97.8125) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:41 loss 1.4293 (1.3184) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 842.711, TIME@all 0.304 -epoch: [99/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 1:03:50 loss 1.3998 (1.2831) acc 93.7500 (97.8125) lr 0.260000 -epoch: [99/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:42 loss 1.2675 (1.3141) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 842.932, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.2244 (1.2805) acc 100.0000 (97.8125) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:03:26 loss 1.2521 (1.3160) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 842.813, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.3322 (1.2754) acc 96.8750 (98.1250) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3548 (1.3122) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.887, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.4706 (1.2786) acc 93.7500 (98.2812) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.2693 (1.3024) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.835, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:03:33 loss 1.3026 (1.3010) acc 96.8750 (97.6562) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3877 (1.3196) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 842.826, TIME@all 0.304 -epoch: [100/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 1:03:32 loss 1.4139 (1.2917) acc 96.8750 (97.8125) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:25 loss 1.3959 (1.3114) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 843.031, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:03:33 loss 1.3819 (1.2840) acc 96.8750 (98.1250) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.4425 (1.3198) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 842.815, TIME@all 0.304 -epoch: [100/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 1:03:32 loss 1.4303 (1.2862) acc 93.7500 (97.6562) lr 0.260000 -epoch: [100/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.3051 (1.3041) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 843.178, TIME@all 0.304 -epoch: [100/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:03:33 loss 1.2208 (1.2729) acc 96.8750 (98.1250) lr 0.260000 -epoch: [100/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:03:26 loss 1.4042 (1.3167) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 842.955, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.011) eta 1:03:17 loss 1.2758 (1.3116) acc 96.8750 (97.3438) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.3988 (1.3174) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 843.106, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2951 (1.3032) acc 96.8750 (97.5000) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2523 (1.3350) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 843.122, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.3241 (1.3297) acc 96.8750 (97.5000) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.3143 (1.3480) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 843.121, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 1:03:17 loss 1.2668 (1.3153) acc 100.0000 (96.8750) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:06 loss 1.2421 (1.3265) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.258, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 1:03:16 loss 1.4025 (1.3410) acc 87.5000 (96.7188) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:03:06 loss 1.4164 (1.3374) acc 90.6250 (96.7188) lr 0.260000 -FPS@all 843.316, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2200 (1.2974) acc 100.0000 (97.8125) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2886 (1.3198) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 843.114, TIME@all 0.304 -epoch: [101/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 1:03:17 loss 1.2578 (1.3023) acc 100.0000 (97.6562) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:03:07 loss 1.2949 (1.3351) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 843.109, TIME@all 0.304 -epoch: [101/350][20/50] time 0.309 (0.304) data 0.000 (0.012) eta 1:03:16 loss 1.2068 (1.3150) acc 100.0000 (96.8750) lr 0.260000 -epoch: [101/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 1:03:06 loss 1.2554 (1.3416) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 843.469, TIME@all 0.304 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:02:53 loss 1.3802 (1.3740) acc 93.7500 (96.0938) lr 0.260000 -epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 1:02:52 loss 1.3824 (1.3697) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 844.004, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 1:02:52 loss 1.3888 (1.3330) acc 96.8750 (96.5625) lr 0.260000 -epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3981 (1.3538) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 844.180, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.6200 (1.3466) acc 90.6250 (97.0312) lr 0.260000 -epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.3783 (1.3780) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 844.018, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.2498 (1.3330) acc 100.0000 (96.5625) lr 0.260000 -epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.3974 (1.3652) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 844.012, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:52 loss 1.3326 (1.3483) acc 96.8750 (96.8750) lr 0.260000 -epoch: [102/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3324 (1.3580) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 844.367, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:52 loss 1.2905 (1.3222) acc 100.0000 (97.3438) lr 0.260000 -epoch: [102/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.2915 (1.3672) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 844.072, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 1:02:52 loss 1.3555 (1.3190) acc 96.8750 (96.8750) lr 0.260000 -epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:51 loss 1.3751 (1.3427) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 844.215, TIME@all 0.303 -epoch: [102/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:53 loss 1.4298 (1.3406) acc 93.7500 (96.5625) lr 0.260000 -epoch: [102/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:52 loss 1.2613 (1.3588) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 844.045, TIME@all 0.303 -epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4078 (1.3129) acc 93.7500 (96.7188) lr 0.260000 -epoch: [103/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.3290 (1.3357) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 844.046, TIME@all 0.303 -epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:40 loss 1.3623 (1.2921) acc 93.7500 (97.8125) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:34 loss 1.2085 (1.3293) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 844.077, TIME@all 0.303 -epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4099 (1.3072) acc 93.7500 (95.7812) lr 0.260000 -epoch: [103/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.2867 (1.3121) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 844.148, TIME@all 0.303 -epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.2363 (1.2844) acc 100.0000 (97.1875) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:33 loss 1.3313 (1.3247) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 844.246, TIME@all 0.303 -epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.3069 (1.2800) acc 96.8750 (98.2812) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:02:34 loss 1.2012 (1.2993) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 844.065, TIME@all 0.303 -epoch: [103/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.3518 (1.3047) acc 96.8750 (97.9688) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 1:02:34 loss 1.3356 (1.3328) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 844.397, TIME@all 0.303 -epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:39 loss 1.4207 (1.2950) acc 93.7500 (97.3438) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.2841 (1.3174) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 844.041, TIME@all 0.303 -epoch: [103/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:02:39 loss 1.4055 (1.3026) acc 96.8750 (97.6562) lr 0.260000 -epoch: [103/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:02:34 loss 1.3186 (1.3310) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 844.181, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:24 loss 1.5898 (1.3247) acc 90.6250 (97.3438) lr 0.260000 -epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:02:17 loss 1.3666 (1.3456) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 844.117, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:23 loss 1.2613 (1.3049) acc 100.0000 (97.9688) lr 0.260000 -epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.3026 (1.3309) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 844.197, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:23 loss 1.2761 (1.3518) acc 100.0000 (96.5625) lr 0.260000 -epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.2783 (1.3441) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 844.158, TIME@all 0.303 -epoch: [104/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:02:24 loss 1.3617 (1.2978) acc 96.8750 (97.1875) lr 0.260000 -epoch: [104/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.3997 (1.3249) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 844.141, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:23 loss 1.4601 (1.3309) acc 93.7500 (96.8750) lr 0.260000 -epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:17 loss 1.4657 (1.3466) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 844.145, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:23 loss 1.4244 (1.3290) acc 93.7500 (95.6250) lr 0.260000 -epoch: [104/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.2894 (1.3421) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 844.285, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 1:02:22 loss 1.2670 (1.3255) acc 93.7500 (96.5625) lr 0.260000 -epoch: [104/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:02:16 loss 1.4765 (1.3560) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 844.345, TIME@all 0.303 -epoch: [104/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 1:02:23 loss 1.5102 (1.3221) acc 96.8750 (97.3438) lr 0.260000 -epoch: [104/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 1:02:16 loss 1.1758 (1.3470) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 844.504, TIME@all 0.303 -epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:08 loss 1.1906 (1.2827) acc 100.0000 (97.8125) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.5185 (1.3250) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 842.200, TIME@all 0.304 -epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.4717 (1.3153) acc 96.8750 (97.6562) lr 0.260000 -epoch: [105/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2903 (1.3278) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 842.275, TIME@all 0.304 -epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.2969 (1.3023) acc 96.8750 (97.3438) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2732 (1.3333) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 842.242, TIME@all 0.304 -epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:08 loss 1.3989 (1.3238) acc 93.7500 (96.5625) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.2817 (1.3189) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.236, TIME@all 0.304 -epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 1:02:07 loss 1.3410 (1.3014) acc 96.8750 (97.6562) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:02:10 loss 1.2710 (1.3314) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 842.254, TIME@all 0.304 -epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 1:02:07 loss 1.3283 (1.2951) acc 96.8750 (97.1875) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:02:10 loss 1.4156 (1.3387) acc 90.6250 (95.9375) lr 0.260000 -FPS@all 842.556, TIME@all 0.304 -epoch: [105/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 1:02:07 loss 1.3207 (1.2740) acc 96.8750 (97.6562) lr 0.260000 -epoch: [105/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:02:09 loss 1.3241 (1.3176) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 842.410, TIME@all 0.304 -epoch: [105/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 1:02:07 loss 1.3284 (1.3071) acc 100.0000 (96.7188) lr 0.260000 -epoch: [105/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:02:09 loss 1.3385 (1.3271) acc 93.7500 (95.8594) lr 0.260000 -FPS@all 842.370, TIME@all 0.304 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.4231 (1.3414) acc 93.7500 (96.4062) lr 0.260000 -epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.3492 (1.3730) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 836.210, TIME@all 0.306 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.5058 (1.3506) acc 93.7500 (96.2500) lr 0.260000 -epoch: [106/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.4318 (1.3918) acc 90.6250 (95.1562) lr 0.260000 -FPS@all 836.191, TIME@all 0.306 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:26 loss 1.5742 (1.3312) acc 87.5000 (97.5000) lr 0.260000 -epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.4296 (1.3698) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 836.302, TIME@all 0.306 -epoch: [106/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.6270 (1.3435) acc 87.5000 (96.2500) lr 0.260000 -epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 1:02:19 loss 1.3431 (1.3529) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 836.419, TIME@all 0.306 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.012) eta 1:02:27 loss 1.4971 (1.3277) acc 93.7500 (96.8750) lr 0.260000 -epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:02:20 loss 1.2449 (1.3682) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 836.220, TIME@all 0.306 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.013) eta 1:02:27 loss 1.5277 (1.3468) acc 93.7500 (97.3438) lr 0.260000 -epoch: [106/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 1:02:20 loss 1.3492 (1.3696) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 836.190, TIME@all 0.306 -epoch: [106/350][20/50] time 0.307 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.3071 (1.3750) acc 100.0000 (96.2500) lr 0.260000 -epoch: [106/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 1:02:19 loss 1.4583 (1.3877) acc 90.6250 (96.0938) lr 0.260000 -FPS@all 836.378, TIME@all 0.306 -epoch: [106/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 1:02:26 loss 1.5378 (1.3741) acc 90.6250 (95.4688) lr 0.260000 -epoch: [106/350][40/50] time 0.310 (0.306) data 0.000 (0.007) eta 1:02:20 loss 1.4589 (1.3943) acc 90.6250 (95.0781) lr 0.260000 -FPS@all 836.564, TIME@all 0.306 -epoch: [107/350][20/50] time 0.304 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.3277 (1.2870) acc 96.8750 (97.6562) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2950 (1.3198) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 839.567, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.2343 (1.3079) acc 100.0000 (97.6562) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3900 (1.3457) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 839.612, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.5857 (1.3503) acc 90.6250 (96.5625) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.4229 (1.3519) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 839.553, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 1:02:03 loss 1.2932 (1.3197) acc 100.0000 (97.0312) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3882 (1.3391) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 839.533, TIME@all 0.305 -epoch: [107/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.4367 (1.3276) acc 93.7500 (96.5625) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2554 (1.3504) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 839.933, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.001 (0.012) eta 1:02:03 loss 1.2361 (1.3216) acc 100.0000 (96.5625) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.2967 (1.3589) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 839.539, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:02:03 loss 1.3800 (1.3323) acc 96.8750 (95.6250) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3175 (1.3528) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 839.676, TIME@all 0.305 -epoch: [107/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:02:02 loss 1.2899 (1.3325) acc 96.8750 (96.5625) lr 0.260000 -epoch: [107/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 1:01:58 loss 1.3748 (1.3334) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 839.719, TIME@all 0.305 -epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.3595 (1.3169) acc 96.8750 (97.5000) lr 0.260000 -epoch: [108/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.4648 (1.3262) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 843.961, TIME@all 0.303 -epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:34 loss 1.3605 (1.3170) acc 100.0000 (97.1875) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.3866 (1.3475) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 844.003, TIME@all 0.303 -epoch: [108/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 1:01:33 loss 1.2888 (1.3212) acc 100.0000 (97.5000) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:01:19 loss 1.2815 (1.3437) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 843.985, TIME@all 0.303 -epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.2441 (1.2857) acc 100.0000 (97.6562) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.5555 (1.3102) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 843.990, TIME@all 0.303 -epoch: [108/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 1:01:33 loss 1.4834 (1.3145) acc 96.8750 (97.1875) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:19 loss 1.3798 (1.3401) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 844.036, TIME@all 0.303 -epoch: [108/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:33 loss 1.2918 (1.2976) acc 100.0000 (97.0312) lr 0.260000 -epoch: [108/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 1:01:18 loss 1.3992 (1.3233) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 844.128, TIME@all 0.303 -epoch: [108/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:32 loss 1.3835 (1.3312) acc 100.0000 (96.7188) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 1:01:18 loss 1.3855 (1.3421) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 844.170, TIME@all 0.303 -epoch: [108/350][20/50] time 0.309 (0.304) data 0.001 (0.013) eta 1:01:33 loss 1.3256 (1.3006) acc 93.7500 (97.1875) lr 0.260000 -epoch: [108/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:01:18 loss 1.3340 (1.3310) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 844.295, TIME@all 0.303 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:17 loss 1.2709 (1.2638) acc 100.0000 (98.2812) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:01:12 loss 1.1685 (1.2716) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 842.729, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:17 loss 1.2369 (1.2489) acc 100.0000 (98.2812) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 1:01:11 loss 1.2754 (1.2625) acc 96.8750 (97.8125) lr 0.260000 -FPS@all 842.817, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.2625 (1.2643) acc 100.0000 (98.2812) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.2364 (1.2655) acc 100.0000 (98.3594) lr 0.260000 -FPS@all 842.798, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:01:16 loss 1.2341 (1.2609) acc 100.0000 (98.4375) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:01:11 loss 1.3706 (1.2726) acc 96.8750 (98.1250) lr 0.260000 -FPS@all 842.968, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.3636 (1.2467) acc 96.8750 (98.9062) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 1:01:11 loss 1.2940 (1.2580) acc 93.7500 (98.4375) lr 0.260000 -FPS@all 842.755, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.2368 (1.2697) acc 100.0000 (98.2812) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:12 loss 1.2415 (1.2691) acc 100.0000 (98.2812) lr 0.260000 -FPS@all 842.770, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 1:01:17 loss 1.2131 (1.2544) acc 100.0000 (98.4375) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.4144 (1.2650) acc 87.5000 (97.9688) lr 0.260000 -FPS@all 842.895, TIME@all 0.304 -epoch: [109/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:01:17 loss 1.1822 (1.2346) acc 100.0000 (98.9062) lr 0.260000 -epoch: [109/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 1:01:11 loss 1.2715 (1.2609) acc 100.0000 (98.2031) lr 0.260000 -FPS@all 843.136, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:00 loss 1.3670 (1.2342) acc 96.8750 (98.5938) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.3073 (1.2839) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 841.269, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:00:59 loss 1.2480 (1.2901) acc 100.0000 (97.3438) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.1989 (1.3129) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 841.320, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:00:59 loss 1.3606 (1.2492) acc 93.7500 (98.2812) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.2446 (1.2872) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 841.294, TIME@all 0.304 -epoch: [110/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 1:01:00 loss 1.6144 (1.2714) acc 90.6250 (98.1250) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 1:00:59 loss 1.3337 (1.3038) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 841.260, TIME@all 0.304 -epoch: [110/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 1:00:59 loss 1.2311 (1.2546) acc 100.0000 (98.9062) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:58 loss 1.3480 (1.3016) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 841.502, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:00:59 loss 1.2340 (1.2585) acc 100.0000 (98.1250) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.001 (0.006) eta 1:00:59 loss 1.2215 (1.2917) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 841.303, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 1:01:00 loss 1.1796 (1.2515) acc 100.0000 (97.9688) lr 0.260000 -epoch: [110/350][40/50] time 0.307 (0.305) data 0.001 (0.006) eta 1:00:59 loss 1.2498 (1.2956) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 841.613, TIME@all 0.304 -epoch: [110/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 1:00:59 loss 1.4523 (1.2628) acc 93.7500 (97.8125) lr 0.260000 -epoch: [110/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:58 loss 1.2610 (1.2987) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 841.437, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2056 (1.2305) acc 100.0000 (98.5938) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:40 loss 1.4500 (1.2646) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 841.646, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2255 (1.2324) acc 100.0000 (98.7500) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:41 loss 1.2874 (1.2585) acc 93.7500 (98.1250) lr 0.260000 -FPS@all 841.539, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.1697 (1.2224) acc 100.0000 (99.0625) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 1:00:41 loss 1.2515 (1.2678) acc 100.0000 (97.9688) lr 0.260000 -FPS@all 841.592, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:49 loss 1.4148 (1.2345) acc 93.7500 (98.4375) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 1:00:40 loss 1.2550 (1.2774) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 841.583, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.2182 (1.2431) acc 96.8750 (98.7500) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:00:40 loss 1.1681 (1.2609) acc 100.0000 (98.4375) lr 0.260000 -FPS@all 841.754, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 1:00:49 loss 1.2645 (1.2582) acc 100.0000 (98.2812) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 1:00:40 loss 1.2473 (1.2818) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 841.585, TIME@all 0.304 -epoch: [111/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.3151 (1.2575) acc 100.0000 (98.2812) lr 0.260000 -epoch: [111/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 1:00:40 loss 1.3016 (1.2731) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 841.697, TIME@all 0.304 -epoch: [111/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:48 loss 1.1676 (1.2415) acc 100.0000 (98.5938) lr 0.260000 -epoch: [111/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 1:00:40 loss 1.2874 (1.2728) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 841.930, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 1:00:45 loss 1.3093 (1.2650) acc 96.8750 (98.2812) lr 0.260000 -epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 1:00:30 loss 1.2168 (1.2925) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 842.171, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.3320 (1.2650) acc 93.7500 (97.8125) lr 0.260000 -epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.2311 (1.2960) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 842.265, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:46 loss 1.3578 (1.2740) acc 96.8750 (97.8125) lr 0.260000 -epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1968 (1.2935) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 842.218, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.2679 (1.2391) acc 100.0000 (98.9062) lr 0.260000 -epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:29 loss 1.1948 (1.2627) acc 100.0000 (97.9688) lr 0.260000 -FPS@all 842.342, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.4564 (1.2545) acc 96.8750 (97.8125) lr 0.260000 -epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:29 loss 1.1948 (1.3019) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 842.387, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.3382 (1.2392) acc 96.8750 (98.5938) lr 0.260000 -epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1653 (1.2728) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 842.157, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.001 (0.013) eta 1:00:45 loss 1.5002 (1.2702) acc 93.7500 (98.1250) lr 0.260000 -epoch: [112/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1963 (1.2911) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 842.193, TIME@all 0.304 -epoch: [112/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 1:00:45 loss 1.4212 (1.2422) acc 93.7500 (98.7500) lr 0.260000 -epoch: [112/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 1:00:30 loss 1.1997 (1.2706) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 842.520, TIME@all 0.304 -epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 1:00:22 loss 1.2995 (1.2885) acc 100.0000 (97.8125) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:00:10 loss 1.3383 (1.2958) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 842.375, TIME@all 0.304 -epoch: [113/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2431 (1.3031) acc 100.0000 (96.4062) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 1:00:10 loss 1.3879 (1.3134) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 842.339, TIME@all 0.304 -epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2704 (1.2903) acc 100.0000 (97.9688) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.4671 (1.3211) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.406, TIME@all 0.304 -epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.3645 (1.2868) acc 90.6250 (97.3438) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.2594 (1.3028) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 842.377, TIME@all 0.304 -epoch: [113/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 1:00:21 loss 1.2271 (1.3004) acc 100.0000 (97.1875) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:09 loss 1.2839 (1.3111) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 842.575, TIME@all 0.304 -epoch: [113/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 1:00:22 loss 1.3390 (1.2867) acc 96.8750 (96.7188) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.2741 (1.3061) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 842.366, TIME@all 0.304 -epoch: [113/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 1:00:21 loss 1.2821 (1.2604) acc 96.8750 (98.2812) lr 0.260000 -epoch: [113/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 1:00:09 loss 1.2677 (1.2934) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 842.538, TIME@all 0.304 -epoch: [113/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 1:00:22 loss 1.2265 (1.2854) acc 100.0000 (97.1875) lr 0.260000 -epoch: [113/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 1:00:10 loss 1.3106 (1.2944) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.702, TIME@all 0.304 -epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:55 loss 1.5116 (1.3313) acc 84.3750 (96.0938) lr 0.260000 -epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.3258 (1.3377) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 843.708, TIME@all 0.303 -epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:59:54 loss 1.5532 (1.3158) acc 90.6250 (96.4062) lr 0.260000 -epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.5179 (1.3413) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 843.774, TIME@all 0.303 -epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.4952 (1.3363) acc 93.7500 (96.2500) lr 0.260000 -epoch: [114/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.2965 (1.3583) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 843.743, TIME@all 0.303 -epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.5596 (1.3110) acc 87.5000 (97.0312) lr 0.260000 -epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:48 loss 1.3721 (1.3414) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 843.928, TIME@all 0.303 -epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:55 loss 1.4324 (1.3432) acc 90.6250 (97.0312) lr 0.260000 -epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:49 loss 1.2720 (1.3354) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 843.737, TIME@all 0.303 -epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:59:54 loss 1.5527 (1.3340) acc 87.5000 (95.4688) lr 0.260000 -epoch: [114/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:59:48 loss 1.3685 (1.3320) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 843.786, TIME@all 0.303 -epoch: [114/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:59:53 loss 1.3550 (1.3133) acc 100.0000 (97.0312) lr 0.260000 -epoch: [114/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:48 loss 1.2578 (1.3259) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 844.078, TIME@all 0.303 -epoch: [114/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:59:54 loss 1.6998 (1.3130) acc 90.6250 (96.8750) lr 0.260000 -epoch: [114/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:48 loss 1.2870 (1.3148) acc 90.6250 (97.1875) lr 0.260000 -FPS@all 843.871, TIME@all 0.303 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.2870 (1.2828) acc 96.8750 (97.9688) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.3474 (1.2936) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 842.980, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.2734 (1.2804) acc 100.0000 (97.9688) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.2895 (1.3160) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.008, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.3843 (1.2994) acc 93.7500 (97.0312) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:36 loss 1.3141 (1.3277) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 842.998, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:39 loss 1.3150 (1.2838) acc 96.8750 (97.6562) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:36 loss 1.3566 (1.3143) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.139, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:59:40 loss 1.3794 (1.2930) acc 96.8750 (96.7188) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:59:37 loss 1.3802 (1.3174) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 842.930, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:40 loss 1.3077 (1.2815) acc 96.8750 (98.4375) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:36 loss 1.2579 (1.3172) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.098, TIME@all 0.304 -epoch: [115/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:59:40 loss 1.3086 (1.2885) acc 100.0000 (97.8125) lr 0.260000 -epoch: [115/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:59:37 loss 1.4217 (1.2918) acc 90.6250 (97.4219) lr 0.260000 -FPS@all 842.950, TIME@all 0.304 -epoch: [115/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:59:39 loss 1.3145 (1.2737) acc 100.0000 (98.5938) lr 0.260000 -epoch: [115/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:36 loss 1.3237 (1.2974) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 843.272, TIME@all 0.304 -epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.3426 (1.2669) acc 96.8750 (98.1250) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.3648 (1.2888) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 843.837, TIME@all 0.303 -epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.2335 (1.2787) acc 100.0000 (97.9688) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.4575 (1.2954) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.937, TIME@all 0.303 -epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.1966 (1.2437) acc 100.0000 (98.5938) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:59:15 loss 1.5126 (1.2966) acc 90.6250 (97.6562) lr 0.260000 -FPS@all 843.903, TIME@all 0.303 -epoch: [116/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:59:20 loss 1.3063 (1.2766) acc 100.0000 (97.6562) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:59:15 loss 1.3393 (1.3039) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 843.903, TIME@all 0.303 -epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:59:20 loss 1.2379 (1.2524) acc 100.0000 (97.9688) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:59:15 loss 1.2761 (1.2916) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 843.883, TIME@all 0.303 -epoch: [116/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:59:19 loss 1.2715 (1.2512) acc 93.7500 (98.2812) lr 0.260000 -epoch: [116/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:14 loss 1.4063 (1.2831) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 844.083, TIME@all 0.303 -epoch: [116/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:59:20 loss 1.2319 (1.2940) acc 96.8750 (97.3438) lr 0.260000 -epoch: [116/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:59:14 loss 1.4744 (1.3153) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 844.022, TIME@all 0.303 -epoch: [116/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:59:22 loss 1.3257 (1.2621) acc 93.7500 (97.3438) lr 0.260000 -epoch: [116/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:59:16 loss 1.2083 (1.2863) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 844.081, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.012) eta 0:59:03 loss 1.2283 (1.2584) acc 100.0000 (97.8125) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:59:01 loss 1.2560 (1.2959) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 843.835, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.4536 (1.2618) acc 93.7500 (98.7500) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:01 loss 1.1758 (1.2819) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 843.878, TIME@all 0.303 -epoch: [117/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:59:02 loss 1.3461 (1.2465) acc 93.7500 (97.9688) lr 0.260000 -epoch: [117/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:59:01 loss 1.2067 (1.2762) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 843.927, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.2918 (1.2771) acc 100.0000 (97.1875) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:59:01 loss 1.2850 (1.2953) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 843.855, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.3944 (1.2401) acc 93.7500 (98.7500) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:01 loss 1.2631 (1.2762) acc 100.0000 (98.0469) lr 0.260000 -FPS@all 843.845, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.001 (0.013) eta 0:59:02 loss 1.2862 (1.2453) acc 100.0000 (99.0625) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:00 loss 1.2983 (1.2888) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 844.064, TIME@all 0.303 -epoch: [117/350][20/50] time 0.306 (0.303) data 0.000 (0.013) eta 0:59:03 loss 1.4304 (1.2740) acc 90.6250 (97.0312) lr 0.260000 -epoch: [117/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:59:00 loss 1.2662 (1.2843) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 843.986, TIME@all 0.303 -epoch: [117/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:59:06 loss 1.3668 (1.2629) acc 96.8750 (98.5938) lr 0.260000 -epoch: [117/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:59:02 loss 1.4280 (1.2978) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 844.048, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.3826 (1.2686) acc 93.7500 (97.8125) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.4601 (1.3266) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 843.614, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2227 (1.2961) acc 96.8750 (96.7188) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.4673 (1.3248) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 843.602, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.4743 (1.3198) acc 90.6250 (96.5625) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.2148 (1.3401) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 843.541, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.4551 (1.2940) acc 96.8750 (97.6562) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3218 (1.3241) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 843.570, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:52 loss 1.2702 (1.2993) acc 100.0000 (97.1875) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:48 loss 1.2692 (1.3359) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 843.754, TIME@all 0.303 -epoch: [118/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2419 (1.3083) acc 100.0000 (97.3438) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3828 (1.3324) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 843.916, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:53 loss 1.2693 (1.2837) acc 96.8750 (97.9688) lr 0.260000 -epoch: [118/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:49 loss 1.3145 (1.3247) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 843.556, TIME@all 0.303 -epoch: [118/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:53 loss 1.3301 (1.3042) acc 100.0000 (96.8750) lr 0.260000 -epoch: [118/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:58:48 loss 1.4125 (1.3400) acc 90.6250 (96.0156) lr 0.260000 -FPS@all 843.701, TIME@all 0.303 -epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:58:45 loss 1.2189 (1.2832) acc 100.0000 (97.6562) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:58:40 loss 1.3084 (1.2955) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 842.016, TIME@all 0.304 -epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2108 (1.2569) acc 100.0000 (98.4375) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.4392 (1.2992) acc 90.6250 (97.3438) lr 0.260000 -FPS@all 842.037, TIME@all 0.304 -epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2346 (1.2726) acc 96.8750 (97.6562) lr 0.260000 -epoch: [119/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:39 loss 1.3725 (1.2920) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.089, TIME@all 0.304 -epoch: [119/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.3277 (1.2816) acc 96.8750 (98.2812) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.2696 (1.3094) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 842.034, TIME@all 0.304 -epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:45 loss 1.3439 (1.2869) acc 93.7500 (97.9688) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:58:40 loss 1.3934 (1.3271) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.024, TIME@all 0.304 -epoch: [119/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.4097 (1.2543) acc 93.7500 (97.6562) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.3380 (1.2899) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.210, TIME@all 0.304 -epoch: [119/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.3419 (1.2608) acc 96.8750 (98.1250) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.2585 (1.2952) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 842.137, TIME@all 0.304 -epoch: [119/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:58:44 loss 1.2080 (1.2846) acc 100.0000 (97.1875) lr 0.260000 -epoch: [119/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:39 loss 1.2368 (1.3090) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 842.457, TIME@all 0.304 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:58:16 loss 1.1914 (1.2792) acc 100.0000 (97.1875) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.2974 (1.3067) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.979, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:58:16 loss 1.2297 (1.2705) acc 100.0000 (98.2812) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.3054 (1.2884) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.930, TIME@all 0.303 -epoch: [120/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:58:17 loss 1.3647 (1.2727) acc 93.7500 (97.9688) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:58:15 loss 1.2677 (1.2892) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 843.918, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.3146 (1.2765) acc 96.8750 (98.2812) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:15 loss 1.3980 (1.3016) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 843.959, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:58:16 loss 1.3402 (1.2999) acc 93.7500 (97.5000) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:14 loss 1.3401 (1.3204) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 844.136, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.2608 (1.2789) acc 100.0000 (97.6562) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:58:15 loss 1.3451 (1.3128) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 843.967, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:16 loss 1.3470 (1.2971) acc 96.8750 (97.1875) lr 0.260000 -epoch: [120/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:58:14 loss 1.4234 (1.3148) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 844.109, TIME@all 0.303 -epoch: [120/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:58:15 loss 1.3630 (1.2733) acc 93.7500 (97.8125) lr 0.260000 -epoch: [120/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:58:15 loss 1.3325 (1.3007) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 844.355, TIME@all 0.303 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.3307 (1.2494) acc 96.8750 (97.9688) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2770 (1.2721) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 842.840, TIME@all 0.304 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.2181 (1.2451) acc 100.0000 (97.5000) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2270 (1.2738) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 842.857, TIME@all 0.304 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.3157 (1.2700) acc 96.8750 (97.5000) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2520 (1.3073) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 842.814, TIME@all 0.304 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:58:16 loss 1.2523 (1.2612) acc 100.0000 (98.1250) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2206 (1.2787) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.004, TIME@all 0.304 -epoch: [121/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.2706 (1.2674) acc 100.0000 (97.9688) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.3208 (1.2901) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 842.773, TIME@all 0.304 -epoch: [121/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:58:17 loss 1.1784 (1.2333) acc 100.0000 (99.0625) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2866 (1.2683) acc 96.8750 (98.1250) lr 0.260000 -FPS@all 842.828, TIME@all 0.304 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:58:16 loss 1.1862 (1.2614) acc 100.0000 (97.8125) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.2084 (1.2928) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 842.957, TIME@all 0.304 -epoch: [121/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:58:16 loss 1.2333 (1.2437) acc 100.0000 (98.9062) lr 0.260000 -epoch: [121/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:58:06 loss 1.3476 (1.2845) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 843.176, TIME@all 0.304 -epoch: [122/350][20/50] time 0.305 (0.304) data 0.001 (0.012) eta 0:57:50 loss 1.3072 (1.2317) acc 100.0000 (98.4375) lr 0.260000 -epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.4415 (1.2919) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 843.283, TIME@all 0.304 -epoch: [122/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:57:51 loss 1.4054 (1.2612) acc 93.7500 (97.3438) lr 0.260000 -epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.3710 (1.2942) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 843.336, TIME@all 0.304 -epoch: [122/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.2822 (1.2384) acc 96.8750 (98.2812) lr 0.260000 -epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.2941 (1.2868) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 843.384, TIME@all 0.304 -epoch: [122/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:57:50 loss 1.2855 (1.2601) acc 96.8750 (97.9688) lr 0.260000 -epoch: [122/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:57:48 loss 1.4240 (1.2888) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 843.502, TIME@all 0.303 -epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:50 loss 1.3622 (1.2367) acc 96.8750 (98.5938) lr 0.260000 -epoch: [122/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:57:49 loss 1.3676 (1.2815) acc 93.7500 (97.8906) lr 0.260000 -FPS@all 843.313, TIME@all 0.304 -epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.3139 (1.2450) acc 93.7500 (98.2812) lr 0.260000 -epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.4563 (1.2813) acc 90.6250 (97.4219) lr 0.260000 -FPS@all 843.294, TIME@all 0.304 -epoch: [122/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:57:50 loss 1.2449 (1.2217) acc 96.8750 (98.9062) lr 0.260000 -epoch: [122/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:57:49 loss 1.5711 (1.2783) acc 90.6250 (97.3438) lr 0.260000 -FPS@all 843.453, TIME@all 0.304 -epoch: [122/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:50 loss 1.2230 (1.2607) acc 100.0000 (98.5938) lr 0.260000 -epoch: [122/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:57:49 loss 1.3406 (1.2784) acc 100.0000 (98.0469) lr 0.260000 -FPS@all 843.599, TIME@all 0.303 -epoch: [123/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:47 loss 1.2898 (1.2845) acc 100.0000 (97.6562) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.3849 (1.3257) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 842.197, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:47 loss 1.3526 (1.2825) acc 93.7500 (98.1250) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:40 loss 1.4097 (1.3212) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.214, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:57:46 loss 1.2299 (1.2789) acc 100.0000 (98.2812) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:39 loss 1.4965 (1.3326) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 842.277, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:47 loss 1.2677 (1.2923) acc 96.8750 (97.6562) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.4258 (1.3466) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 842.207, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:46 loss 1.2708 (1.2991) acc 96.8750 (97.5000) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:39 loss 1.4756 (1.3435) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 842.400, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:57:47 loss 1.2200 (1.2841) acc 100.0000 (97.8125) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:40 loss 1.4220 (1.3326) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 842.193, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:57:47 loss 1.1678 (1.2728) acc 100.0000 (97.9688) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:57:40 loss 1.3882 (1.3073) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.481, TIME@all 0.304 -epoch: [123/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:46 loss 1.1637 (1.2799) acc 100.0000 (97.3438) lr 0.260000 -epoch: [123/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:57:39 loss 1.3679 (1.3214) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 842.339, TIME@all 0.304 -epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:32 loss 1.5447 (1.3529) acc 90.6250 (96.5625) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:57:23 loss 1.3545 (1.3664) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 841.915, TIME@all 0.304 -epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.4949 (1.3566) acc 93.7500 (95.7812) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:23 loss 1.2417 (1.3966) acc 100.0000 (95.3906) lr 0.260000 -FPS@all 841.960, TIME@all 0.304 -epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:57:32 loss 1.4692 (1.3580) acc 93.7500 (96.2500) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:57:23 loss 1.3137 (1.3787) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 841.925, TIME@all 0.304 -epoch: [124/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:57:33 loss 1.4357 (1.3764) acc 93.7500 (96.0938) lr 0.260000 -epoch: [124/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:57:23 loss 1.4052 (1.3976) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 841.990, TIME@all 0.304 -epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:57:31 loss 1.3650 (1.3395) acc 96.8750 (96.5625) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:22 loss 1.3434 (1.3789) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 842.144, TIME@all 0.304 -epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.7716 (1.3599) acc 87.5000 (96.5625) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:22 loss 1.2920 (1.3968) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 842.082, TIME@all 0.304 -epoch: [124/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:57:34 loss 1.4759 (1.3512) acc 100.0000 (96.7188) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:57:24 loss 1.2690 (1.3788) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 842.124, TIME@all 0.304 -epoch: [124/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:57:32 loss 1.5004 (1.3540) acc 93.7500 (96.0938) lr 0.260000 -epoch: [124/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:57:23 loss 1.3394 (1.3831) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 841.939, TIME@all 0.304 -epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:10 loss 1.5197 (1.2731) acc 96.8750 (98.5938) lr 0.260000 -epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:57:07 loss 1.3869 (1.3123) acc 90.6250 (96.9531) lr 0.260000 -FPS@all 842.726, TIME@all 0.304 -epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:57:10 loss 1.3971 (1.2799) acc 93.7500 (97.6562) lr 0.260000 -epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:57:07 loss 1.2835 (1.3019) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 842.677, TIME@all 0.304 -epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.5004 (1.2926) acc 93.7500 (97.6562) lr 0.260000 -epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.4063 (1.3086) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.731, TIME@all 0.304 -epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:10 loss 1.5023 (1.3104) acc 90.6250 (97.0312) lr 0.260000 -epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.3727 (1.3137) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 842.692, TIME@all 0.304 -epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:57:09 loss 1.3305 (1.2718) acc 96.8750 (98.4375) lr 0.260000 -epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2658 (1.2836) acc 96.8750 (98.2031) lr 0.260000 -FPS@all 842.871, TIME@all 0.304 -epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:10 loss 1.2700 (1.2598) acc 96.8750 (98.5938) lr 0.260000 -epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:07 loss 1.2065 (1.2850) acc 100.0000 (98.2031) lr 0.260000 -FPS@all 842.673, TIME@all 0.304 -epoch: [125/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.3521 (1.2769) acc 96.8750 (98.1250) lr 0.260000 -epoch: [125/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2957 (1.2999) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 842.829, TIME@all 0.304 -epoch: [125/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:57:09 loss 1.2638 (1.2567) acc 100.0000 (98.5938) lr 0.260000 -epoch: [125/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:57:06 loss 1.2769 (1.2979) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.111, TIME@all 0.304 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:54 loss 1.3301 (1.2852) acc 100.0000 (97.1875) lr 0.260000 -epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.4258 (1.3075) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 843.675, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2665 (1.2653) acc 96.8750 (98.5938) lr 0.260000 -epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3919 (1.2846) acc 93.7500 (98.1250) lr 0.260000 -FPS@all 843.726, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2953 (1.2903) acc 100.0000 (96.8750) lr 0.260000 -epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3128 (1.2956) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 843.719, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.2856 (1.2762) acc 96.8750 (97.5000) lr 0.260000 -epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.4890 (1.2903) acc 87.5000 (97.4219) lr 0.260000 -FPS@all 843.712, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:56:53 loss 1.2817 (1.2703) acc 100.0000 (97.9688) lr 0.260000 -epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:45 loss 1.2119 (1.2812) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.912, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3765 (1.2845) acc 100.0000 (97.3438) lr 0.260000 -epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:45 loss 1.3231 (1.2970) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 843.862, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3433 (1.2737) acc 93.7500 (98.1250) lr 0.260000 -epoch: [126/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.3107 (1.2946) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.734, TIME@all 0.303 -epoch: [126/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:56:53 loss 1.3048 (1.2715) acc 100.0000 (97.8125) lr 0.260000 -epoch: [126/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:56:46 loss 1.2328 (1.2991) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 843.990, TIME@all 0.303 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.2252 (1.2455) acc 100.0000 (98.4375) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.2098 (1.2662) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 842.702, TIME@all 0.304 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.4037 (1.2544) acc 96.8750 (98.1250) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3082 (1.2783) acc 96.8750 (98.1250) lr 0.260000 -FPS@all 842.764, TIME@all 0.304 -epoch: [127/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.2958 (1.2441) acc 96.8750 (98.2812) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3749 (1.2828) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.706, TIME@all 0.304 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.3346 (1.2550) acc 96.8750 (98.4375) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.3842 (1.2832) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 842.728, TIME@all 0.304 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:56:38 loss 1.4448 (1.2742) acc 84.3750 (97.0312) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:56:36 loss 1.2842 (1.2813) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 842.899, TIME@all 0.304 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:39 loss 1.2574 (1.2799) acc 100.0000 (97.9688) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:37 loss 1.2237 (1.2901) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 842.704, TIME@all 0.304 -epoch: [127/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.3448 (1.2336) acc 93.7500 (98.2812) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:56:36 loss 1.3340 (1.2750) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.851, TIME@all 0.304 -epoch: [127/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:56:38 loss 1.3450 (1.2406) acc 93.7500 (98.4375) lr 0.260000 -epoch: [127/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:56:36 loss 1.2658 (1.2842) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.127, TIME@all 0.304 -epoch: [128/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.3010 (1.2609) acc 96.8750 (98.1250) lr 0.260000 -epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.5283 (1.3024) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 843.855, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.2592 (1.2672) acc 96.8750 (97.9688) lr 0.260000 -epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.4982 (1.3007) acc 87.5000 (96.6406) lr 0.260000 -FPS@all 843.774, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:30 loss 1.4899 (1.2694) acc 93.7500 (97.5000) lr 0.260000 -epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.2180 (1.2923) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 843.837, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.1819 (1.2489) acc 100.0000 (98.2812) lr 0.260000 -epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.6716 (1.2875) acc 90.6250 (97.5000) lr 0.260000 -FPS@all 843.999, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:56:29 loss 1.2873 (1.2548) acc 96.8750 (97.6562) lr 0.260000 -epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.006) eta 0:56:18 loss 1.3470 (1.2921) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 843.813, TIME@all 0.303 -epoch: [128/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2148 (1.2571) acc 96.8750 (97.9688) lr 0.260000 -epoch: [128/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.4824 (1.2898) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 844.056, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2150 (1.2338) acc 100.0000 (98.1250) lr 0.260000 -epoch: [128/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:56:17 loss 1.3850 (1.2884) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 843.961, TIME@all 0.303 -epoch: [128/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:56:29 loss 1.2279 (1.2479) acc 100.0000 (98.7500) lr 0.260000 -epoch: [128/350][40/50] time 0.310 (0.304) data 0.000 (0.007) eta 0:56:18 loss 1.3729 (1.2891) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 843.795, TIME@all 0.303 -epoch: [129/350][20/50] time 0.307 (0.304) data 0.001 (0.014) eta 0:56:07 loss 1.3831 (1.3302) acc 93.7500 (96.5625) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.4259 (1.3338) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 843.472, TIME@all 0.304 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:07 loss 1.3206 (1.3045) acc 100.0000 (97.5000) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:56:00 loss 1.5771 (1.3303) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 843.477, TIME@all 0.304 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:07 loss 1.2487 (1.3046) acc 100.0000 (97.3438) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.3360 (1.3132) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 843.435, TIME@all 0.304 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:06 loss 1.2470 (1.3045) acc 96.8750 (97.5000) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.4581 (1.3265) acc 90.6250 (96.9531) lr 0.260000 -FPS@all 843.488, TIME@all 0.304 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:06 loss 1.3081 (1.3383) acc 93.7500 (96.4062) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:59 loss 1.3840 (1.3534) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 843.623, TIME@all 0.303 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:56:06 loss 1.4072 (1.3167) acc 90.6250 (97.0312) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.5290 (1.3467) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 843.442, TIME@all 0.304 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:56:06 loss 1.3488 (1.3219) acc 96.8750 (97.0312) lr 0.260000 -epoch: [129/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:59 loss 1.5394 (1.3534) acc 90.6250 (96.0938) lr 0.260000 -FPS@all 843.656, TIME@all 0.303 -epoch: [129/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:56:06 loss 1.2727 (1.3208) acc 100.0000 (97.5000) lr 0.260000 -epoch: [129/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:56:00 loss 1.3048 (1.3335) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 843.810, TIME@all 0.303 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.011) eta 0:56:02 loss 1.3151 (1.3025) acc 96.8750 (97.6562) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:51 loss 1.4262 (1.3331) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 841.721, TIME@all 0.304 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:56:01 loss 1.3418 (1.3078) acc 96.8750 (96.7188) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:50 loss 1.2080 (1.3171) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 841.794, TIME@all 0.304 -epoch: [130/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:56:02 loss 1.3225 (1.3061) acc 96.8750 (97.0312) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:55:51 loss 1.2693 (1.3151) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 841.720, TIME@all 0.304 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2633 (1.3451) acc 96.8750 (96.2500) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:50 loss 1.3988 (1.3520) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 841.937, TIME@all 0.304 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2827 (1.2966) acc 96.8750 (97.8125) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:51 loss 1.4064 (1.3066) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 841.751, TIME@all 0.304 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.001 (0.012) eta 0:56:00 loss 1.2067 (1.2980) acc 100.0000 (97.0312) lr 0.260000 -epoch: [130/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:55:50 loss 1.3743 (1.3084) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 842.162, TIME@all 0.304 -epoch: [130/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.4782 (1.3249) acc 90.6250 (97.1875) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:50 loss 1.2729 (1.3214) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 841.880, TIME@all 0.304 -epoch: [130/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:56:01 loss 1.2626 (1.2688) acc 100.0000 (98.1250) lr 0.260000 -epoch: [130/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:55:51 loss 1.3367 (1.3020) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 841.733, TIME@all 0.304 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.011) eta 0:56:08 loss 1.3230 (1.2933) acc 93.7500 (96.5625) lr 0.260000 -epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3873 (1.3023) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 840.382, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:56:07 loss 1.4502 (1.2986) acc 96.8750 (97.3438) lr 0.260000 -epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.4301 (1.3253) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 840.480, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.001 (0.013) eta 0:56:07 loss 1.3671 (1.3041) acc 96.8750 (97.1875) lr 0.260000 -epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:55:45 loss 1.2571 (1.3261) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 840.622, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:56:08 loss 1.3987 (1.2724) acc 96.8750 (98.5938) lr 0.260000 -epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3235 (1.3125) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 840.440, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:56:08 loss 1.2898 (1.2953) acc 100.0000 (98.5938) lr 0.260000 -epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.3192 (1.3035) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 840.417, TIME@all 0.305 -epoch: [131/350][20/50] time 0.303 (0.307) data 0.000 (0.012) eta 0:56:08 loss 1.4391 (1.2971) acc 93.7500 (97.6562) lr 0.260000 -epoch: [131/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.2889 (1.3130) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 840.668, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:56:07 loss 1.3273 (1.3153) acc 96.8750 (96.8750) lr 0.260000 -epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:45 loss 1.3891 (1.3282) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 840.573, TIME@all 0.305 -epoch: [131/350][20/50] time 0.304 (0.307) data 0.001 (0.012) eta 0:56:08 loss 1.3400 (1.2797) acc 93.7500 (97.8125) lr 0.260000 -epoch: [131/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:55:46 loss 1.2739 (1.3050) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 840.436, TIME@all 0.305 -epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.5788 (1.2956) acc 93.7500 (97.6562) lr 0.260000 -epoch: [132/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.2804 (1.3407) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 841.942, TIME@all 0.304 -epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:55:28 loss 1.6019 (1.3219) acc 90.6250 (97.5000) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:55:20 loss 1.2238 (1.3444) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 841.844, TIME@all 0.304 -epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:55:27 loss 1.3400 (1.3129) acc 100.0000 (96.8750) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.1957 (1.3268) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 842.020, TIME@all 0.304 -epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.4667 (1.3281) acc 90.6250 (96.5625) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.3598 (1.3527) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 841.859, TIME@all 0.304 -epoch: [132/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:55:27 loss 1.5164 (1.3193) acc 87.5000 (96.7188) lr 0.260000 -epoch: [132/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.4052 (1.3712) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 842.062, TIME@all 0.304 -epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:27 loss 1.4561 (1.3456) acc 93.7500 (96.7188) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.2597 (1.3521) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 841.855, TIME@all 0.304 -epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:28 loss 1.4913 (1.3118) acc 90.6250 (97.1875) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:55:20 loss 1.2528 (1.3466) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 841.839, TIME@all 0.304 -epoch: [132/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:55:28 loss 1.4151 (1.3236) acc 96.8750 (96.5625) lr 0.260000 -epoch: [132/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:55:20 loss 1.3242 (1.3492) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 842.163, TIME@all 0.304 -epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:55:12 loss 1.4685 (1.2909) acc 84.3750 (97.1875) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:55:07 loss 1.3594 (1.3224) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 842.208, TIME@all 0.304 -epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.2974 (1.2730) acc 96.8750 (97.9688) lr 0.260000 -epoch: [133/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3994 (1.3152) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.260, TIME@all 0.304 -epoch: [133/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.3856 (1.3053) acc 93.7500 (97.3438) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3608 (1.3235) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 842.248, TIME@all 0.304 -epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:11 loss 1.3169 (1.3368) acc 96.8750 (96.4062) lr 0.260000 -epoch: [133/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 0:55:07 loss 1.3026 (1.3335) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 842.232, TIME@all 0.304 -epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:12 loss 1.4077 (1.3005) acc 93.7500 (96.5625) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.2074 (1.3139) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 842.231, TIME@all 0.304 -epoch: [133/350][20/50] time 0.309 (0.304) data 0.001 (0.014) eta 0:55:11 loss 1.2615 (1.2862) acc 96.8750 (96.8750) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:55:06 loss 1.2492 (1.3042) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 842.433, TIME@all 0.304 -epoch: [133/350][20/50] time 0.309 (0.304) data 0.000 (0.013) eta 0:55:11 loss 1.2177 (1.2891) acc 100.0000 (97.9688) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:06 loss 1.4572 (1.3269) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.376, TIME@all 0.304 -epoch: [133/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:55:11 loss 1.3593 (1.3016) acc 90.6250 (97.3438) lr 0.260000 -epoch: [133/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:55:07 loss 1.3031 (1.3112) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 842.556, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:48 loss 1.3410 (1.2629) acc 96.8750 (98.7500) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.1839 (1.2956) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 843.122, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:54:47 loss 1.2577 (1.2678) acc 100.0000 (98.2812) lr 0.260000 -epoch: [134/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.2745 (1.2743) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 843.166, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:54:47 loss 1.2855 (1.2779) acc 96.8750 (98.2812) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.1966 (1.2881) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 843.072, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3113 (1.2574) acc 100.0000 (98.2812) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.4087 (1.2866) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.007, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:54:46 loss 1.2016 (1.2576) acc 100.0000 (98.2812) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:45 loss 1.2294 (1.2644) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 843.209, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3256 (1.2512) acc 96.8750 (97.9688) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:46 loss 1.2105 (1.2693) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 843.004, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.4061 (1.2646) acc 96.8750 (98.7500) lr 0.260000 -epoch: [134/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:54:45 loss 1.2318 (1.2754) acc 100.0000 (98.2812) lr 0.260000 -FPS@all 843.236, TIME@all 0.304 -epoch: [134/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:47 loss 1.3654 (1.2820) acc 96.8750 (97.1875) lr 0.260000 -epoch: [134/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:54:46 loss 1.3079 (1.3012) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 843.417, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4973 (1.2762) acc 96.8750 (97.9688) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:33 loss 1.2814 (1.2954) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 842.593, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4306 (1.2951) acc 100.0000 (96.7188) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.2207 (1.2873) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 842.626, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.3652 (1.2919) acc 100.0000 (97.9688) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.2149 (1.3093) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.679, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.5105 (1.3126) acc 96.8750 (96.5625) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.3157 (1.3288) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 842.636, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:32 loss 1.2794 (1.2786) acc 100.0000 (97.8125) lr 0.260000 -epoch: [135/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:54:32 loss 1.3445 (1.2980) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 842.837, TIME@all 0.304 -epoch: [135/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:33 loss 1.4925 (1.3187) acc 87.5000 (97.5000) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:54:33 loss 1.2806 (1.3166) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 842.631, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:54:33 loss 1.4290 (1.3088) acc 96.8750 (96.7188) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:54:32 loss 1.5640 (1.3185) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 842.989, TIME@all 0.304 -epoch: [135/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:33 loss 1.5069 (1.2775) acc 96.8750 (98.1250) lr 0.260000 -epoch: [135/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:54:32 loss 1.3361 (1.2906) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 842.779, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:28 loss 1.4885 (1.2563) acc 96.8750 (98.1250) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.3504 (1.2888) acc 90.6250 (96.7969) lr 0.260000 -FPS@all 843.019, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:28 loss 1.3968 (1.2489) acc 93.7500 (97.6562) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.2249 (1.2677) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 843.019, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:28 loss 1.2718 (1.2608) acc 100.0000 (98.7500) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:16 loss 1.2357 (1.2879) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 843.064, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:54:27 loss 1.2671 (1.2506) acc 100.0000 (98.7500) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:15 loss 1.2627 (1.2741) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.234, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:28 loss 1.3539 (1.2565) acc 93.7500 (97.9688) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:16 loss 1.2493 (1.2817) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 843.030, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:54:27 loss 1.3545 (1.2553) acc 96.8750 (97.9688) lr 0.260000 -epoch: [136/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:54:16 loss 1.3209 (1.2908) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 843.095, TIME@all 0.304 -epoch: [136/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:54:27 loss 1.2738 (1.2745) acc 96.8750 (97.6562) lr 0.260000 -epoch: [136/350][40/50] time 0.298 (0.304) data 0.001 (0.007) eta 0:54:15 loss 1.3795 (1.2979) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 843.388, TIME@all 0.304 -epoch: [136/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:54:27 loss 1.3901 (1.2698) acc 87.5000 (97.5000) lr 0.260000 -epoch: [136/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:54:15 loss 1.3081 (1.2990) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 843.180, TIME@all 0.304 -epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:54:01 loss 1.3013 (1.2419) acc 93.7500 (97.6562) lr 0.260000 -epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:56 loss 1.3243 (1.2913) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 844.897, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:01 loss 1.3665 (1.2781) acc 93.7500 (97.6562) lr 0.260000 -epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:56 loss 1.4143 (1.3098) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 844.936, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:01 loss 1.1968 (1.2439) acc 96.8750 (98.2812) lr 0.260000 -epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:56 loss 1.2938 (1.2718) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 844.997, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:54:01 loss 1.3618 (1.2529) acc 96.8750 (98.4375) lr 0.260000 -epoch: [137/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:53:57 loss 1.3405 (1.2839) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 844.924, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:54:01 loss 1.3276 (1.2534) acc 96.8750 (98.4375) lr 0.260000 -epoch: [137/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:53:57 loss 1.2611 (1.2868) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 844.922, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:01 loss 1.3107 (1.2499) acc 96.8750 (98.1250) lr 0.260000 -epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:53:57 loss 1.2723 (1.2876) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 845.245, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:54:01 loss 1.2676 (1.2583) acc 100.0000 (97.5000) lr 0.260000 -epoch: [137/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:56 loss 1.4476 (1.2888) acc 84.3750 (97.1875) lr 0.260000 -FPS@all 845.105, TIME@all 0.303 -epoch: [137/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:54:00 loss 1.1975 (1.2634) acc 100.0000 (98.4375) lr 0.260000 -epoch: [137/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:53:56 loss 1.2043 (1.2909) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 845.084, TIME@all 0.303 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.4468 (1.3041) acc 93.7500 (98.1250) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.3801 (1.3294) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 842.757, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:53:52 loss 1.2151 (1.3108) acc 100.0000 (97.6562) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:53:48 loss 1.3431 (1.3306) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 842.706, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.2374 (1.3028) acc 100.0000 (96.4062) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.1926 (1.3151) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 842.731, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:52 loss 1.2963 (1.3125) acc 96.8750 (96.7188) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:48 loss 1.2661 (1.3374) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 842.748, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:53:51 loss 1.1919 (1.3064) acc 100.0000 (97.5000) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:47 loss 1.4500 (1.3444) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 842.954, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:50 loss 1.3882 (1.3031) acc 96.8750 (97.8125) lr 0.260000 -epoch: [138/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:53:47 loss 1.2686 (1.3283) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 842.960, TIME@all 0.304 -epoch: [138/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:53:51 loss 1.2739 (1.2964) acc 100.0000 (97.6562) lr 0.260000 -epoch: [138/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:53:47 loss 1.4284 (1.3238) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 843.145, TIME@all 0.304 -epoch: [138/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:53:52 loss 1.3631 (1.2925) acc 93.7500 (97.6562) lr 0.260000 -epoch: [138/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:53:48 loss 1.3716 (1.3240) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 842.883, TIME@all 0.304 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:53:32 loss 1.2804 (1.2813) acc 93.7500 (97.6562) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:28 loss 1.2219 (1.2864) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 844.142, TIME@all 0.303 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.2189 (1.2718) acc 100.0000 (97.9688) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:27 loss 1.4054 (1.3016) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 844.173, TIME@all 0.303 -epoch: [139/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.4167 (1.2809) acc 90.6250 (97.3438) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:53:28 loss 1.2979 (1.2921) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 844.186, TIME@all 0.303 -epoch: [139/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.3132 (1.2718) acc 93.7500 (98.2812) lr 0.260000 -epoch: [139/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:53:28 loss 1.3859 (1.2876) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 844.152, TIME@all 0.303 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:53:32 loss 1.3456 (1.2534) acc 100.0000 (98.2812) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.3660 (1.2698) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 844.359, TIME@all 0.303 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.2762 (1.2623) acc 96.8750 (98.2812) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:28 loss 1.3713 (1.2910) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 844.162, TIME@all 0.303 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.3446 (1.2559) acc 96.8750 (97.9688) lr 0.260000 -epoch: [139/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.5127 (1.2874) acc 87.5000 (97.4219) lr 0.260000 -FPS@all 844.550, TIME@all 0.303 -epoch: [139/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:53:32 loss 1.4574 (1.2841) acc 93.7500 (97.0312) lr 0.260000 -epoch: [139/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:53:27 loss 1.4456 (1.2921) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 844.311, TIME@all 0.303 -epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.011) eta 0:53:33 loss 1.3497 (1.2610) acc 96.8750 (98.1250) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.2169 (1.2794) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 842.093, TIME@all 0.304 -epoch: [140/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.3054 (1.2613) acc 100.0000 (98.5938) lr 0.260000 -epoch: [140/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 0:53:20 loss 1.2882 (1.2691) acc 96.8750 (98.2031) lr 0.260000 -FPS@all 842.211, TIME@all 0.304 -epoch: [140/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.4066 (1.2836) acc 96.8750 (98.1250) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.1786 (1.2829) acc 100.0000 (97.9688) lr 0.260000 -FPS@all 842.101, TIME@all 0.304 -epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:53:32 loss 1.2868 (1.2683) acc 100.0000 (98.2812) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:53:20 loss 1.2208 (1.2822) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 842.144, TIME@all 0.304 -epoch: [140/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 0:53:32 loss 1.3662 (1.2902) acc 100.0000 (96.8750) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:53:20 loss 1.3515 (1.2887) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 842.116, TIME@all 0.304 -epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:53:32 loss 1.3697 (1.2700) acc 93.7500 (97.5000) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:53:19 loss 1.2047 (1.2730) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 842.301, TIME@all 0.304 -epoch: [140/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:53:32 loss 1.3897 (1.2696) acc 96.8750 (98.1250) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:53:20 loss 1.1839 (1.2732) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 842.252, TIME@all 0.304 -epoch: [140/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:53:30 loss 1.3367 (1.2681) acc 96.8750 (98.9062) lr 0.260000 -epoch: [140/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:53:19 loss 1.3236 (1.2716) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 842.641, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:53:07 loss 1.3378 (1.2846) acc 100.0000 (97.8125) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:58 loss 1.7600 (1.3056) acc 84.3750 (97.0312) lr 0.260000 -FPS@all 843.200, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.3527 (1.2600) acc 96.8750 (97.8125) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4446 (1.3154) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 843.261, TIME@all 0.304 -epoch: [141/350][20/50] time 0.309 (0.304) data 0.000 (0.012) eta 0:53:06 loss 1.3650 (1.2747) acc 96.8750 (97.0312) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:58 loss 1.3453 (1.3018) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 843.275, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.4077 (1.2831) acc 93.7500 (98.2812) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4448 (1.3049) acc 90.6250 (97.3438) lr 0.260000 -FPS@all 843.199, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.3782 (1.2639) acc 93.7500 (97.9688) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.4685 (1.2923) acc 90.6250 (97.5000) lr 0.260000 -FPS@all 843.192, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.2522 (1.2732) acc 96.8750 (97.9688) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:57 loss 1.3590 (1.3030) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.402, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:06 loss 1.3373 (1.2694) acc 100.0000 (97.9688) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.5739 (1.3055) acc 87.5000 (96.7969) lr 0.260000 -FPS@all 843.350, TIME@all 0.304 -epoch: [141/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:53:07 loss 1.2546 (1.2472) acc 100.0000 (97.9688) lr 0.260000 -epoch: [141/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:58 loss 1.5873 (1.2970) acc 90.6250 (96.7969) lr 0.260000 -FPS@all 843.485, TIME@all 0.304 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:52:47 loss 1.3005 (1.3046) acc 96.8750 (97.0312) lr 0.260000 -epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:52:41 loss 1.3749 (1.2980) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 843.916, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:52:47 loss 1.3140 (1.2776) acc 90.6250 (97.1875) lr 0.260000 -epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:52:41 loss 1.2937 (1.2783) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 843.939, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2443 (1.2536) acc 96.8750 (98.2812) lr 0.260000 -epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.2564 (1.2779) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 843.957, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.4135 (1.2860) acc 96.8750 (97.8125) lr 0.260000 -epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.3137 (1.2883) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 843.935, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2621 (1.2668) acc 96.8750 (98.5938) lr 0.260000 -epoch: [142/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:40 loss 1.1941 (1.2912) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 844.106, TIME@all 0.303 -epoch: [142/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2294 (1.2779) acc 96.8750 (97.8125) lr 0.260000 -epoch: [142/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.3028 (1.2910) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 843.950, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2008 (1.2680) acc 100.0000 (97.5000) lr 0.260000 -epoch: [142/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:41 loss 1.1670 (1.2790) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 844.089, TIME@all 0.303 -epoch: [142/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:52:47 loss 1.2659 (1.2670) acc 96.8750 (97.9688) lr 0.260000 -epoch: [142/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:52:41 loss 1.3597 (1.2864) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 844.297, TIME@all 0.303 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:52:38 loss 1.2547 (1.2643) acc 96.8750 (97.3438) lr 0.260000 -epoch: [143/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:31 loss 1.3084 (1.3038) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 842.927, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.2349 (1.2564) acc 96.8750 (97.8125) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2230 (1.2854) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 842.977, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2960 (1.2776) acc 96.8750 (97.6562) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2353 (1.2989) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 843.021, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.014) eta 0:52:37 loss 1.3392 (1.2860) acc 100.0000 (97.9688) lr 0.260000 -epoch: [143/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2458 (1.3078) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 843.143, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2509 (1.2689) acc 96.8750 (98.4375) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.3058 (1.2996) acc 90.6250 (97.4219) lr 0.260000 -FPS@all 842.945, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.3878 (1.2852) acc 93.7500 (97.1875) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2037 (1.2973) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 842.911, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:37 loss 1.2633 (1.2418) acc 100.0000 (98.1250) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2229 (1.2585) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 843.081, TIME@all 0.304 -epoch: [143/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:52:38 loss 1.2147 (1.2524) acc 100.0000 (97.9688) lr 0.260000 -epoch: [143/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:31 loss 1.2126 (1.2982) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 843.311, TIME@all 0.304 -epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:52:26 loss 1.2046 (1.2459) acc 100.0000 (98.5938) lr 0.260000 -epoch: [144/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.3119 (1.2837) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.192, TIME@all 0.304 -epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:52:26 loss 1.2397 (1.2685) acc 100.0000 (97.6562) lr 0.260000 -epoch: [144/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.2591 (1.2780) acc 96.8750 (97.8125) lr 0.260000 -FPS@all 843.270, TIME@all 0.304 -epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.2295 (1.2388) acc 96.8750 (98.9062) lr 0.260000 -epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:15 loss 1.3125 (1.2844) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 843.238, TIME@all 0.304 -epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.1786 (1.2522) acc 100.0000 (98.1250) lr 0.260000 -epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:15 loss 1.3192 (1.2845) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 843.178, TIME@all 0.304 -epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:52:25 loss 1.3183 (1.2619) acc 96.8750 (97.8125) lr 0.260000 -epoch: [144/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:52:14 loss 1.3035 (1.3094) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 843.400, TIME@all 0.304 -epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.3332 (1.2713) acc 96.8750 (97.8125) lr 0.260000 -epoch: [144/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:15 loss 1.2593 (1.2839) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 843.232, TIME@all 0.304 -epoch: [144/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:52:25 loss 1.1654 (1.2632) acc 100.0000 (97.5000) lr 0.260000 -epoch: [144/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:52:14 loss 1.2565 (1.2890) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 843.358, TIME@all 0.304 -epoch: [144/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:52:26 loss 1.2032 (1.2430) acc 100.0000 (98.9062) lr 0.260000 -epoch: [144/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:52:14 loss 1.2541 (1.2743) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 843.560, TIME@all 0.303 -epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.4142 (1.3455) acc 96.8750 (95.6250) lr 0.260000 -epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4686 (1.3423) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 842.557, TIME@all 0.304 -epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:05 loss 1.5974 (1.3207) acc 93.7500 (96.7188) lr 0.260000 -epoch: [145/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4272 (1.3399) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 842.626, TIME@all 0.304 -epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:52:06 loss 1.3909 (1.3285) acc 93.7500 (96.8750) lr 0.260000 -epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:52:00 loss 1.5932 (1.3530) acc 87.5000 (96.2500) lr 0.260000 -FPS@all 842.600, TIME@all 0.304 -epoch: [145/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:52:05 loss 1.3390 (1.2996) acc 96.8750 (97.5000) lr 0.260000 -epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:59 loss 1.3974 (1.3342) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 842.772, TIME@all 0.304 -epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.4634 (1.3178) acc 96.8750 (97.3438) lr 0.260000 -epoch: [145/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.6142 (1.3392) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 842.603, TIME@all 0.304 -epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:06 loss 1.3526 (1.3423) acc 96.8750 (95.3125) lr 0.260000 -epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.4392 (1.3519) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 842.577, TIME@all 0.304 -epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:52:05 loss 1.4485 (1.3457) acc 96.8750 (96.5625) lr 0.260000 -epoch: [145/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:52:00 loss 1.3340 (1.3589) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 842.898, TIME@all 0.304 -epoch: [145/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:52:05 loss 1.3380 (1.3118) acc 96.8750 (97.0312) lr 0.260000 -epoch: [145/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:51:59 loss 1.4189 (1.3398) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 842.722, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:47 loss 1.4851 (1.2992) acc 93.7500 (96.8750) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:48 loss 1.3972 (1.3363) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 842.487, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:46 loss 1.5751 (1.3134) acc 90.6250 (96.7188) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:47 loss 1.2241 (1.3421) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 842.574, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.6488 (1.2923) acc 87.5000 (97.1875) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.2787 (1.3196) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 842.504, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.3483 (1.2864) acc 90.6250 (97.3438) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.3993 (1.3147) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 842.519, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.4727 (1.2816) acc 93.7500 (97.6562) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.2159 (1.3247) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 842.520, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:46 loss 1.3292 (1.2780) acc 93.7500 (97.5000) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:47 loss 1.2315 (1.3380) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 842.667, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:51:46 loss 1.5755 (1.3187) acc 93.7500 (96.4062) lr 0.260000 -epoch: [146/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:47 loss 1.2692 (1.3715) acc 100.0000 (95.0781) lr 0.260000 -FPS@all 842.731, TIME@all 0.304 -epoch: [146/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:47 loss 1.4629 (1.2978) acc 93.7500 (97.1875) lr 0.260000 -epoch: [146/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:51:48 loss 1.4002 (1.3155) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 842.821, TIME@all 0.304 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.3242 (1.2781) acc 93.7500 (97.5000) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:51:29 loss 1.2298 (1.2905) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 843.550, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.2286 (1.2728) acc 100.0000 (97.3438) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:51:30 loss 1.3611 (1.3113) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 843.501, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:51:38 loss 1.3143 (1.2755) acc 100.0000 (97.8125) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:51:29 loss 1.2684 (1.3022) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 843.605, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:38 loss 1.2725 (1.2707) acc 100.0000 (97.3438) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:29 loss 1.2980 (1.2998) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 843.650, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:51:38 loss 1.3720 (1.2705) acc 96.8750 (97.3438) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:51:30 loss 1.2759 (1.3047) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 843.532, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:39 loss 1.3223 (1.2836) acc 93.7500 (97.1875) lr 0.260000 -epoch: [147/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:51:30 loss 1.2614 (1.2944) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 843.536, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:38 loss 1.3368 (1.2621) acc 93.7500 (98.2812) lr 0.260000 -epoch: [147/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:51:29 loss 1.2227 (1.2987) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 843.939, TIME@all 0.303 -epoch: [147/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:51:38 loss 1.3327 (1.2642) acc 100.0000 (99.2188) lr 0.260000 -epoch: [147/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:51:29 loss 1.2200 (1.2868) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 843.694, TIME@all 0.303 -epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:20 loss 1.3826 (1.2724) acc 90.6250 (97.3438) lr 0.260000 -epoch: [148/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3571 (1.2875) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 844.688, TIME@all 0.303 -epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:20 loss 1.4701 (1.2619) acc 90.6250 (98.1250) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3022 (1.2906) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 844.643, TIME@all 0.303 -epoch: [148/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:19 loss 1.3454 (1.2777) acc 96.8750 (97.9688) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3518 (1.3155) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 844.714, TIME@all 0.303 -epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:19 loss 1.2432 (1.2407) acc 96.8750 (98.5938) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3764 (1.2729) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 844.842, TIME@all 0.303 -epoch: [148/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:51:20 loss 1.3008 (1.2571) acc 96.8750 (98.4375) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:51:09 loss 1.3254 (1.2804) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 844.642, TIME@all 0.303 -epoch: [148/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:51:20 loss 1.3215 (1.2474) acc 100.0000 (98.4375) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:09 loss 1.3757 (1.2817) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 844.646, TIME@all 0.303 -epoch: [148/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:51:19 loss 1.4479 (1.2805) acc 93.7500 (96.7188) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.3789 (1.2902) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 845.045, TIME@all 0.303 -epoch: [148/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:51:19 loss 1.2521 (1.2636) acc 100.0000 (98.1250) lr 0.260000 -epoch: [148/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:51:08 loss 1.2954 (1.2999) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 844.772, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:05 loss 1.2033 (1.2302) acc 100.0000 (98.4375) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:50:56 loss 1.3614 (1.2655) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 843.965, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:51:05 loss 1.2975 (1.2570) acc 96.8750 (97.9688) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:50:56 loss 1.3808 (1.2745) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 843.999, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.1885 (1.2205) acc 100.0000 (99.0625) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:50:56 loss 1.3008 (1.2652) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 844.078, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:51:04 loss 1.2248 (1.2544) acc 100.0000 (98.4375) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.2481 (1.2697) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 844.186, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.2117 (1.2629) acc 100.0000 (98.5938) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:50:56 loss 1.2864 (1.2939) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 844.005, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.2307 (1.2376) acc 96.8750 (98.2812) lr 0.260000 -epoch: [149/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3132 (1.2689) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 844.350, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:04 loss 1.3027 (1.2624) acc 96.8750 (98.2812) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3052 (1.2851) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 844.131, TIME@all 0.303 -epoch: [149/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:51:05 loss 1.1968 (1.2651) acc 100.0000 (97.8125) lr 0.260000 -epoch: [149/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:56 loss 1.3394 (1.2827) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 843.985, TIME@all 0.303 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:50:55 loss 1.2840 (1.2479) acc 96.8750 (98.2812) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.3233 (1.2834) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 842.873, TIME@all 0.304 -epoch: [150/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:50:54 loss 1.2961 (1.2648) acc 96.8750 (98.2812) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.2387 (1.2994) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 842.949, TIME@all 0.304 -epoch: [150/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:50:55 loss 1.3931 (1.2728) acc 90.6250 (97.5000) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.4986 (1.3017) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 842.866, TIME@all 0.304 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:55 loss 1.2101 (1.2501) acc 100.0000 (98.5938) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:46 loss 1.3625 (1.2889) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 842.873, TIME@all 0.304 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:50:55 loss 1.5794 (1.2983) acc 90.6250 (97.3438) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 0:50:46 loss 1.2514 (1.3188) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 842.882, TIME@all 0.304 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:54 loss 1.2837 (1.2239) acc 96.8750 (98.5938) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:45 loss 1.1729 (1.2521) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 843.062, TIME@all 0.304 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:54 loss 1.4042 (1.2425) acc 93.7500 (98.4375) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:50:45 loss 1.3278 (1.2879) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 843.046, TIME@all 0.304 -epoch: [150/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:55 loss 1.3825 (1.2691) acc 96.8750 (97.5000) lr 0.260000 -epoch: [150/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:50:46 loss 1.3972 (1.2942) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 843.157, TIME@all 0.304 -epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:50:42 loss 1.1923 (1.2009) acc 100.0000 (99.2188) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.2724 (1.2144) acc 96.8750 (98.5938) lr 0.026000 -FPS@all 840.796, TIME@all 0.304 -epoch: [151/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:50:42 loss 1.1842 (1.1968) acc 96.8750 (99.2188) lr 0.026000 -epoch: [151/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.2199 (1.1989) acc 96.8750 (98.9062) lr 0.026000 -FPS@all 840.840, TIME@all 0.304 -epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:50:42 loss 1.1701 (1.2133) acc 100.0000 (98.9062) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.1742 (1.2046) acc 96.8750 (98.7500) lr 0.026000 -FPS@all 840.865, TIME@all 0.304 -epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:42 loss 1.2024 (1.2002) acc 100.0000 (99.0625) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:50:35 loss 1.2133 (1.2159) acc 96.8750 (98.2812) lr 0.026000 -FPS@all 840.836, TIME@all 0.304 -epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.2158 (1.1949) acc 96.8750 (98.7500) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:50:35 loss 1.1654 (1.1991) acc 100.0000 (98.7500) lr 0.026000 -FPS@all 841.031, TIME@all 0.304 -epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:50:42 loss 1.1677 (1.1974) acc 100.0000 (98.4375) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:50:35 loss 1.1279 (1.2000) acc 100.0000 (98.5156) lr 0.026000 -FPS@all 840.802, TIME@all 0.304 -epoch: [151/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.1618 (1.1910) acc 100.0000 (99.5312) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:50:35 loss 1.2328 (1.2101) acc 96.8750 (98.7500) lr 0.026000 -FPS@all 840.992, TIME@all 0.304 -epoch: [151/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:50:41 loss 1.1792 (1.1944) acc 100.0000 (99.0625) lr 0.026000 -epoch: [151/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:50:35 loss 1.1189 (1.1909) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 841.250, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:50:23 loss 1.2449 (1.1516) acc 100.0000 (99.8438) lr 0.026000 -epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:50:16 loss 1.1544 (1.1637) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 843.377, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.1719 (1.1470) acc 100.0000 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1322 (1.1484) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.414, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:22 loss 1.3518 (1.1615) acc 96.8750 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1401 (1.1577) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.445, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:23 loss 1.1718 (1.1454) acc 100.0000 (99.5312) lr 0.026000 -epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1207 (1.1498) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 843.399, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.2056 (1.1524) acc 100.0000 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1298 (1.1566) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 843.394, TIME@all 0.304 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:22 loss 1.1467 (1.1524) acc 100.0000 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1692 (1.1561) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 843.590, TIME@all 0.303 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:50:23 loss 1.1412 (1.1553) acc 100.0000 (99.3750) lr 0.026000 -epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:16 loss 1.1523 (1.1611) acc 100.0000 (99.1406) lr 0.026000 -FPS@all 843.683, TIME@all 0.303 -epoch: [152/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:50:22 loss 1.2479 (1.1536) acc 96.8750 (99.0625) lr 0.026000 -epoch: [152/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:50:15 loss 1.1481 (1.1493) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 843.552, TIME@all 0.303 -epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:49:56 loss 1.1679 (1.1261) acc 96.8750 (99.8438) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1455 (1.1398) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 844.693, TIME@all 0.303 -epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:49:55 loss 1.1036 (1.1223) acc 100.0000 (100.0000) lr 0.026000 -epoch: [153/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1069 (1.1369) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 844.713, TIME@all 0.303 -epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:49:55 loss 1.1114 (1.1344) acc 100.0000 (99.5312) lr 0.026000 -epoch: [153/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:54 loss 1.1620 (1.1454) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 844.740, TIME@all 0.303 -epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:49:55 loss 1.1030 (1.1306) acc 100.0000 (99.6875) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.1125 (1.1326) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 844.718, TIME@all 0.303 -epoch: [153/350][20/50] time 0.302 (0.303) data 0.001 (0.013) eta 0:49:55 loss 1.1335 (1.1231) acc 100.0000 (99.8438) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.1625 (1.1312) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 844.728, TIME@all 0.303 -epoch: [153/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:49:56 loss 1.2230 (1.1339) acc 96.8750 (99.5312) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:54 loss 1.2110 (1.1499) acc 100.0000 (99.2188) lr 0.026000 -FPS@all 844.953, TIME@all 0.303 -epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:49:55 loss 1.1162 (1.1204) acc 100.0000 (99.8438) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:49:53 loss 1.1904 (1.1307) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 844.929, TIME@all 0.303 -epoch: [153/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:49:55 loss 1.1435 (1.1249) acc 100.0000 (100.0000) lr 0.026000 -epoch: [153/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:53 loss 1.1257 (1.1394) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 844.879, TIME@all 0.303 -epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:49:46 loss 1.1424 (1.1158) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:41 loss 1.1761 (1.1358) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 843.874, TIME@all 0.303 -epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:49:47 loss 1.1103 (1.1228) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:41 loss 1.1205 (1.1254) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.797, TIME@all 0.303 -epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:49:46 loss 1.1088 (1.1198) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:49:40 loss 1.1180 (1.1221) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.875, TIME@all 0.303 -epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:47 loss 1.1326 (1.1238) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:41 loss 1.1363 (1.1299) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.803, TIME@all 0.303 -epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1852 (1.1185) acc 96.8750 (99.5312) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.2492 (1.1303) acc 96.8750 (99.4531) lr 0.026000 -FPS@all 844.031, TIME@all 0.303 -epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1272 (1.1178) acc 100.0000 (100.0000) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:41 loss 1.2030 (1.1312) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 843.829, TIME@all 0.303 -epoch: [154/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:49:47 loss 1.0971 (1.1152) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.1150 (1.1282) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 844.179, TIME@all 0.303 -epoch: [154/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:46 loss 1.1072 (1.1160) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:40 loss 1.1274 (1.1249) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.982, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:49:29 loss 1.1423 (1.1230) acc 100.0000 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:49:22 loss 1.1196 (1.1349) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 845.019, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.2167 (1.1197) acc 96.8750 (99.8438) lr 0.026000 -epoch: [155/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1264 (1.1328) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 845.064, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.1419 (1.1064) acc 100.0000 (100.0000) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1134 (1.1365) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 845.054, TIME@all 0.303 -epoch: [155/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:49:30 loss 1.1795 (1.1107) acc 100.0000 (100.0000) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1544 (1.1314) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 845.038, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:49:29 loss 1.2041 (1.1208) acc 93.7500 (99.5312) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:49:21 loss 1.1249 (1.1339) acc 100.0000 (99.1406) lr 0.026000 -FPS@all 845.209, TIME@all 0.303 -epoch: [155/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:49:29 loss 1.1658 (1.1224) acc 100.0000 (99.5312) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.0982 (1.1326) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 845.339, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:49:30 loss 1.1683 (1.1199) acc 100.0000 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:49:22 loss 1.1818 (1.1326) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 844.994, TIME@all 0.303 -epoch: [155/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:49:29 loss 1.1710 (1.1260) acc 100.0000 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:49:22 loss 1.1338 (1.1392) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 845.143, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.1388 (1.1041) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.0903 (1.1148) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 844.811, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.0997 (1.1083) acc 100.0000 (99.6875) lr 0.026000 -epoch: [156/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1464 (1.1216) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 844.798, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:49:14 loss 1.0892 (1.1119) acc 100.0000 (99.8438) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:49:07 loss 1.1542 (1.1208) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 844.920, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:15 loss 1.0902 (1.1108) acc 100.0000 (99.8438) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1338 (1.1193) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 844.798, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:49:14 loss 1.0917 (1.1034) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:49:06 loss 1.2658 (1.1229) acc 96.8750 (99.3750) lr 0.026000 -FPS@all 845.028, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.0883 (1.1031) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1403 (1.1231) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 844.816, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.1078 (1.1099) acc 100.0000 (99.6875) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1519 (1.1287) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 845.143, TIME@all 0.303 -epoch: [156/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:49:14 loss 1.1398 (1.1063) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:49:07 loss 1.1048 (1.1210) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 844.966, TIME@all 0.303 -epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:49:15 loss 1.0952 (1.1021) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 0:49:01 loss 1.1013 (1.1104) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.042, TIME@all 0.304 -epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0910 (1.1019) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:49:01 loss 1.0947 (1.1149) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.100, TIME@all 0.304 -epoch: [157/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0975 (1.1028) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.1159 (1.1144) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 842.100, TIME@all 0.304 -epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.0913 (1.1063) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:49:01 loss 1.0961 (1.1102) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.093, TIME@all 0.304 -epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:49:14 loss 1.0940 (1.0987) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.1129 (1.1132) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.239, TIME@all 0.304 -epoch: [157/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:49:15 loss 1.1015 (1.1005) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:01 loss 1.0950 (1.1039) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.355, TIME@all 0.304 -epoch: [157/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:49:15 loss 1.0975 (1.1017) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:49:01 loss 1.1030 (1.1161) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 842.116, TIME@all 0.304 -epoch: [157/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:49:14 loss 1.1272 (1.1040) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:49:00 loss 1.1044 (1.1132) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.227, TIME@all 0.304 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.2468 (1.1108) acc 96.8750 (99.8438) lr 0.026000 -epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1951 (1.1185) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.842, TIME@all 0.303 -epoch: [158/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.3219 (1.1146) acc 96.8750 (99.6875) lr 0.026000 -epoch: [158/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1032 (1.1164) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.939, TIME@all 0.303 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.2199 (1.1019) acc 100.0000 (100.0000) lr 0.026000 -epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1025 (1.1072) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.897, TIME@all 0.303 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.3578 (1.1123) acc 90.6250 (99.5312) lr 0.026000 -epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.0821 (1.1149) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.890, TIME@all 0.303 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:48:46 loss 1.1550 (1.1084) acc 100.0000 (99.8438) lr 0.026000 -epoch: [158/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:48:40 loss 1.0950 (1.1179) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 844.074, TIME@all 0.303 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:48:47 loss 1.1667 (1.1056) acc 96.8750 (99.5312) lr 0.026000 -epoch: [158/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:48:40 loss 1.1141 (1.1199) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 843.866, TIME@all 0.303 -epoch: [158/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:48:47 loss 1.1677 (1.1059) acc 100.0000 (99.8438) lr 0.026000 -epoch: [158/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:48:40 loss 1.1022 (1.1131) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 844.217, TIME@all 0.303 -epoch: [158/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:48:46 loss 1.1440 (1.1087) acc 100.0000 (99.6875) lr 0.026000 -epoch: [158/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:48:40 loss 1.1037 (1.1139) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 844.026, TIME@all 0.303 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:48:34 loss 1.1167 (1.1005) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:48:27 loss 1.1698 (1.1063) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 841.592, TIME@all 0.304 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:34 loss 1.1097 (1.0974) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:27 loss 1.1405 (1.1057) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.621, TIME@all 0.304 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:33 loss 1.1292 (1.1000) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:26 loss 1.2004 (1.1066) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 841.640, TIME@all 0.304 -epoch: [159/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1180 (1.1029) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:27 loss 1.1126 (1.1047) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.589, TIME@all 0.304 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1537 (1.0994) acc 100.0000 (99.6875) lr 0.026000 -epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:26 loss 1.1439 (1.1058) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.806, TIME@all 0.304 -epoch: [159/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:48:33 loss 1.0820 (1.1003) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:48:26 loss 1.1452 (1.1067) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.079, TIME@all 0.304 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:48:34 loss 1.1184 (1.0980) acc 100.0000 (99.8438) lr 0.026000 -epoch: [159/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:48:27 loss 1.1392 (1.1083) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.594, TIME@all 0.304 -epoch: [159/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:48:33 loss 1.1169 (1.1005) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:48:26 loss 1.1118 (1.1088) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.763, TIME@all 0.304 -epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.0881 (1.0885) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0969 (1.1051) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.116, TIME@all 0.304 -epoch: [160/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.1250 (1.0910) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0953 (1.1030) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.136, TIME@all 0.304 -epoch: [160/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:48:20 loss 1.0853 (1.0937) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0844 (1.1121) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 841.280, TIME@all 0.304 -epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.0951 (1.0944) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0857 (1.1029) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.145, TIME@all 0.304 -epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:48:20 loss 1.1011 (1.0987) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0823 (1.1053) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.125, TIME@all 0.304 -epoch: [160/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:48:19 loss 1.1079 (1.0981) acc 100.0000 (99.5312) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:48:18 loss 1.0864 (1.1092) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 841.305, TIME@all 0.304 -epoch: [160/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 0:48:20 loss 1.0964 (1.0913) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.0816 (1.1050) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.360, TIME@all 0.304 -epoch: [160/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:48:20 loss 1.0947 (1.0909) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:48:18 loss 1.1002 (1.1068) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.166, TIME@all 0.304 -epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.011) eta 0:48:12 loss 1.0931 (1.0967) acc 100.0000 (100.0000) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0871 (1.1057) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.134, TIME@all 0.305 -epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.1237 (1.1011) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0957 (1.1076) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.144, TIME@all 0.305 -epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0829 (1.1026) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0899 (1.1100) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 839.167, TIME@all 0.305 -epoch: [161/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:48:12 loss 1.0779 (1.0963) acc 100.0000 (99.6875) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0982 (1.1095) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.127, TIME@all 0.305 -epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0956 (1.0948) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0746 (1.1049) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.120, TIME@all 0.305 -epoch: [161/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:48:11 loss 1.0978 (1.1007) acc 100.0000 (99.5312) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:48:10 loss 1.0876 (1.1070) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.316, TIME@all 0.305 -epoch: [161/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:48:12 loss 1.0844 (1.1007) acc 100.0000 (100.0000) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.1241 (1.1150) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.258, TIME@all 0.305 -epoch: [161/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:48:11 loss 1.1024 (1.0966) acc 100.0000 (100.0000) lr 0.026000 -epoch: [161/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:48:10 loss 1.0808 (1.1063) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.413, TIME@all 0.305 -epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:48:07 loss 1.1145 (1.0934) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1645 (1.1066) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.068, TIME@all 0.305 -epoch: [162/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1203 (1.0902) acc 100.0000 (100.0000) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1372 (1.1020) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.179, TIME@all 0.305 -epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1127 (1.1077) acc 100.0000 (99.3750) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:47:55 loss 1.1637 (1.1080) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.146, TIME@all 0.305 -epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.1217 (1.0922) acc 100.0000 (100.0000) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1429 (1.1078) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.074, TIME@all 0.305 -epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:48:06 loss 1.1982 (1.0945) acc 96.8750 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.001 (0.006) eta 0:47:55 loss 1.1126 (1.0952) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.106, TIME@all 0.305 -epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.2037 (1.1017) acc 100.0000 (99.6875) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1161 (1.1092) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.305, TIME@all 0.305 -epoch: [162/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:48:06 loss 1.1093 (1.0876) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:47:55 loss 1.1119 (1.0999) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.247, TIME@all 0.305 -epoch: [162/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:48:04 loss 1.1124 (1.0939) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:47:54 loss 1.0978 (1.0979) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.619, TIME@all 0.305 -epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.011) eta 0:47:46 loss 1.0867 (1.0918) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:38 loss 1.1471 (1.1028) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.011, TIME@all 0.305 -epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:46 loss 1.1066 (1.0977) acc 100.0000 (99.6875) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:38 loss 1.0729 (1.1040) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.014, TIME@all 0.305 -epoch: [163/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1279 (1.0992) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:38 loss 1.0827 (1.1032) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.059, TIME@all 0.305 -epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:45 loss 1.0851 (1.0906) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:37 loss 1.1376 (1.1007) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.247, TIME@all 0.305 -epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1130 (1.1116) acc 100.0000 (99.5312) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1071 (1.1154) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.068, TIME@all 0.305 -epoch: [163/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:47:46 loss 1.1322 (1.1159) acc 100.0000 (99.2188) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:47:37 loss 1.1002 (1.1169) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 840.172, TIME@all 0.305 -epoch: [163/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.1530 (1.1080) acc 100.0000 (99.5312) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1007 (1.1175) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 840.004, TIME@all 0.305 -epoch: [163/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:47:46 loss 1.0855 (1.0955) acc 100.0000 (99.8438) lr 0.026000 -epoch: [163/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:47:37 loss 1.1165 (1.1144) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 840.342, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1784 (1.0929) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1743 (1.1061) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 838.984, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.012) eta 0:47:38 loss 1.1773 (1.0879) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.315 (0.306) data 0.000 (0.006) eta 0:47:28 loss 1.0898 (1.1083) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 838.895, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.001 (0.013) eta 0:47:37 loss 1.0977 (1.0922) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1716 (1.1070) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.960, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.2041 (1.0983) acc 96.8750 (99.5312) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:27 loss 1.1205 (1.1057) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.072, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.2038 (1.1023) acc 96.8750 (99.6875) lr 0.026000 -epoch: [164/350][40/50] time 0.315 (0.306) data 0.000 (0.007) eta 0:47:28 loss 1.1168 (1.1093) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 838.928, TIME@all 0.305 -epoch: [164/350][20/50] time 0.311 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1105 (1.0909) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.317 (0.306) data 0.001 (0.007) eta 0:47:28 loss 1.1198 (1.1036) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.222, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.013) eta 0:47:37 loss 1.1525 (1.0911) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.306) data 0.001 (0.007) eta 0:47:28 loss 1.1748 (1.1091) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 838.954, TIME@all 0.305 -epoch: [164/350][20/50] time 0.316 (0.306) data 0.000 (0.014) eta 0:47:37 loss 1.1291 (1.0868) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.306) data 0.000 (0.007) eta 0:47:27 loss 1.1019 (1.0987) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.118, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.306) data 0.001 (0.012) eta 0:47:15 loss 1.1491 (1.0942) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1605 (1.1041) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.736, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:47:14 loss 1.1877 (1.0938) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1661 (1.1040) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 838.816, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:47:14 loss 1.2130 (1.1018) acc 100.0000 (99.8438) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:07 loss 1.1020 (1.1036) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.917, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.012) eta 0:47:14 loss 1.1748 (1.1103) acc 96.8750 (99.5312) lr 0.026000 -epoch: [165/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1165 (1.1071) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.748, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.306) data 0.001 (0.013) eta 0:47:15 loss 1.2215 (1.0994) acc 96.8750 (99.8438) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.1276 (1.1044) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.747, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.306) data 0.000 (0.013) eta 0:47:15 loss 1.2500 (1.0975) acc 100.0000 (99.6875) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.1472 (1.1050) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.771, TIME@all 0.305 -epoch: [165/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:47:14 loss 1.2347 (1.0985) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:47:08 loss 1.1389 (1.1017) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 838.892, TIME@all 0.305 -epoch: [165/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:47:17 loss 1.2298 (1.0932) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:47:08 loss 1.0918 (1.1002) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.085, TIME@all 0.305 -epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:46:48 loss 1.1262 (1.0909) acc 100.0000 (99.6875) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:47 loss 1.0959 (1.0985) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.350, TIME@all 0.305 -epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:46:47 loss 1.1228 (1.1005) acc 100.0000 (99.8438) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.1519 (1.1087) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.451, TIME@all 0.305 -epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1104 (1.0949) acc 100.0000 (99.8438) lr 0.026000 -epoch: [166/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.1347 (1.1087) acc 96.8750 (99.4531) lr 0.026000 -FPS@all 840.538, TIME@all 0.305 -epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1610 (1.0939) acc 96.8750 (99.8438) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.0716 (1.1095) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 840.387, TIME@all 0.305 -epoch: [166/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:46:47 loss 1.1080 (1.0973) acc 100.0000 (99.6875) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.1966 (1.1024) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 840.436, TIME@all 0.305 -epoch: [166/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:46:48 loss 1.1534 (1.0921) acc 96.8750 (99.8438) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:46:46 loss 1.0937 (1.1037) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.389, TIME@all 0.305 -epoch: [166/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:46:47 loss 1.1113 (1.0889) acc 100.0000 (100.0000) lr 0.026000 -epoch: [166/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:46:46 loss 1.0982 (1.1063) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.580, TIME@all 0.305 -epoch: [166/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:46:49 loss 1.1144 (1.0958) acc 100.0000 (100.0000) lr 0.026000 -epoch: [166/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:46:47 loss 1.1254 (1.1113) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.605, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:42 loss 1.0719 (1.0874) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.0977 (1.0986) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.965, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:46:41 loss 1.0804 (1.0844) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:46:38 loss 1.0963 (1.0968) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.069, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:46:41 loss 1.1135 (1.0868) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:46:39 loss 1.1296 (1.0972) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 838.950, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.1075 (1.0908) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.1278 (1.1017) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.976, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0941 (1.0884) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:39 loss 1.0885 (1.0997) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.994, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0891 (1.0861) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0963 (1.0971) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.311, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0792 (1.0870) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0841 (1.0978) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.116, TIME@all 0.305 -epoch: [167/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:46:41 loss 1.0773 (1.0833) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:46:38 loss 1.0666 (1.0948) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.170, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.1378 (1.0963) acc 100.0000 (99.3750) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0695 (1.0993) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 839.554, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.1198 (1.0930) acc 100.0000 (100.0000) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0955 (1.0968) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.512, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.0802 (1.0915) acc 100.0000 (99.6875) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.1026 (1.1002) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.582, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1020 (1.0942) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0829 (1.1059) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 839.701, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1132 (1.0854) acc 100.0000 (100.0000) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0747 (1.0946) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.552, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:46:24 loss 1.1289 (1.0933) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.2008 (1.1116) acc 96.8750 (99.4531) lr 0.026000 -FPS@all 839.718, TIME@all 0.305 -epoch: [168/350][20/50] time 0.303 (0.305) data 0.001 (0.013) eta 0:46:22 loss 1.1536 (1.0944) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.1337 (1.1056) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 840.005, TIME@all 0.305 -epoch: [168/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:46:24 loss 1.0894 (1.0886) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:46:21 loss 1.0891 (1.0973) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.560, TIME@all 0.305 -epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.013) eta 0:46:27 loss 1.0998 (1.0928) acc 100.0000 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0767 (1.1001) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.146, TIME@all 0.306 -epoch: [169/350][20/50] time 0.307 (0.307) data 0.000 (0.012) eta 0:46:27 loss 1.1573 (1.0861) acc 96.8750 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:46:16 loss 1.1870 (1.0983) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 837.078, TIME@all 0.306 -epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.1018 (1.0839) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0929 (1.0941) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.127, TIME@all 0.306 -epoch: [169/350][20/50] time 0.307 (0.307) data 0.000 (0.013) eta 0:46:27 loss 1.0693 (1.0858) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.298 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.0952 (1.0964) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.060, TIME@all 0.306 -epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.0695 (1.0761) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:16 loss 1.1129 (1.0953) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.118, TIME@all 0.306 -epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:26 loss 1.0952 (1.0832) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:15 loss 1.0939 (1.0918) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.270, TIME@all 0.306 -epoch: [169/350][20/50] time 0.305 (0.307) data 0.001 (0.014) eta 0:46:28 loss 1.0936 (1.0807) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.303 (0.307) data 0.001 (0.007) eta 0:46:17 loss 1.1271 (1.0974) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 837.257, TIME@all 0.306 -epoch: [169/350][20/50] time 0.306 (0.307) data 0.000 (0.014) eta 0:46:27 loss 1.0717 (1.0797) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:46:15 loss 1.1639 (1.0968) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.239, TIME@all 0.306 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.012) eta 0:46:12 loss 1.1162 (1.0818) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:45:57 loss 1.0990 (1.0992) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 837.766, TIME@all 0.306 -epoch: [170/350][20/50] time 0.318 (0.307) data 0.000 (0.013) eta 0:46:13 loss 1.1452 (1.0869) acc 96.8750 (99.6875) lr 0.026000 -epoch: [170/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1139 (1.0972) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.802, TIME@all 0.306 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.012) eta 0:46:11 loss 1.1059 (1.0835) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:45:57 loss 1.1138 (1.0954) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.868, TIME@all 0.306 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:11 loss 1.1278 (1.0845) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.001 (0.007) eta 0:45:56 loss 1.0907 (1.0969) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 837.971, TIME@all 0.305 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:12 loss 1.1013 (1.0827) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.0953 (1.1009) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.801, TIME@all 0.306 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:12 loss 1.0962 (1.0833) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.0701 (1.0950) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.801, TIME@all 0.306 -epoch: [170/350][20/50] time 0.317 (0.307) data 0.000 (0.013) eta 0:46:11 loss 1.1262 (1.0788) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1872 (1.0989) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 837.941, TIME@all 0.306 -epoch: [170/350][20/50] time 0.313 (0.307) data 0.000 (0.013) eta 0:46:10 loss 1.1227 (1.0890) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:45:57 loss 1.1616 (1.0984) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 838.142, TIME@all 0.305 -epoch: [171/350][20/50] time 0.302 (0.305) data 0.001 (0.012) eta 0:45:35 loss 1.1189 (1.0858) acc 100.0000 (100.0000) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.1045 (1.1018) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.831, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:36 loss 1.1699 (1.0942) acc 96.8750 (99.6875) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0770 (1.1079) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 839.786, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.0826 (1.0909) acc 100.0000 (99.6875) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.1174 (1.1070) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.753, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.0856 (1.0823) acc 100.0000 (100.0000) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0766 (1.1044) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.787, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:35 loss 1.1195 (1.0864) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.0900 (1.1030) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.950, TIME@all 0.305 -epoch: [171/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:45:36 loss 1.1203 (1.0843) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:45:34 loss 1.0788 (1.0991) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.066, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:35 loss 1.1141 (1.0858) acc 100.0000 (99.6875) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.1147 (1.1006) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.912, TIME@all 0.305 -epoch: [171/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:45:36 loss 1.1432 (1.0887) acc 100.0000 (100.0000) lr 0.026000 -epoch: [171/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:45:34 loss 1.0842 (1.1099) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.774, TIME@all 0.305 -epoch: [172/350][20/50] time 0.293 (0.304) data 0.000 (0.012) eta 0:45:14 loss 1.1218 (1.0887) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:45:20 loss 1.1246 (1.0988) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.852, TIME@all 0.305 -epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1020 (1.0889) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:21 loss 1.1253 (1.1008) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.781, TIME@all 0.305 -epoch: [172/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:45:17 loss 1.1049 (1.0969) acc 100.0000 (99.6875) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.001 (0.007) eta 0:45:20 loss 1.0831 (1.0963) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.769, TIME@all 0.305 -epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1310 (1.0907) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:21 loss 1.1488 (1.0994) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.753, TIME@all 0.305 -epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.0892 (1.0879) acc 100.0000 (99.8438) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:20 loss 1.1144 (1.0949) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.956, TIME@all 0.305 -epoch: [172/350][20/50] time 0.293 (0.304) data 0.000 (0.012) eta 0:45:14 loss 1.1775 (1.0923) acc 96.8750 (99.6875) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:45:21 loss 1.1429 (1.0999) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 839.779, TIME@all 0.305 -epoch: [172/350][20/50] time 0.294 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.0977 (1.0822) acc 100.0000 (99.8438) lr 0.026000 -epoch: [172/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:20 loss 1.1172 (1.0915) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.901, TIME@all 0.305 -epoch: [172/350][20/50] time 0.295 (0.304) data 0.000 (0.013) eta 0:45:14 loss 1.1533 (1.0874) acc 96.8750 (99.6875) lr 0.026000 -epoch: [172/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:45:20 loss 1.1105 (1.1049) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.058, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1816 (1.0811) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1557 (1.0985) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.688, TIME@all 0.305 -epoch: [173/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:45:05 loss 1.2135 (1.0828) acc 96.8750 (99.8438) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1162 (1.0934) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.707, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.0829 (1.0802) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.0891 (1.0958) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.763, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.0905 (1.0780) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:01 loss 1.1273 (1.0958) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.912, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.1001 (1.0830) acc 100.0000 (99.8438) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1184 (1.0976) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.676, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.014) eta 0:45:03 loss 1.1035 (1.0846) acc 100.0000 (99.8438) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1540 (1.0963) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.840, TIME@all 0.305 -epoch: [173/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1219 (1.0823) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1681 (1.1005) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.687, TIME@all 0.305 -epoch: [173/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:45:03 loss 1.1594 (1.0793) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:45:02 loss 1.1221 (1.0975) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.098, TIME@all 0.305 -epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:44:52 loss 1.0816 (1.0831) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:44:52 loss 1.1398 (1.0978) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.094, TIME@all 0.305 -epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0836 (1.0806) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:51 loss 1.0993 (1.0946) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.145, TIME@all 0.305 -epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:44:52 loss 1.0747 (1.0822) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.306 (0.306) data 0.001 (0.006) eta 0:44:51 loss 1.1181 (1.0923) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.158, TIME@all 0.305 -epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.1088 (1.0859) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:44:52 loss 1.0770 (1.0983) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.108, TIME@all 0.305 -epoch: [174/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0823 (1.0867) acc 100.0000 (99.5312) lr 0.026000 -epoch: [174/350][40/50] time 0.307 (0.306) data 0.001 (0.007) eta 0:44:51 loss 1.1231 (1.0967) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 839.123, TIME@all 0.305 -epoch: [174/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:44:48 loss 1.0921 (1.0811) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:44:50 loss 1.0759 (1.0923) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.768, TIME@all 0.305 -epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0777 (1.0831) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:44:51 loss 1.1997 (1.1017) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 839.311, TIME@all 0.305 -epoch: [174/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:44:52 loss 1.0720 (1.0876) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.307 (0.306) data 0.001 (0.007) eta 0:44:51 loss 1.1266 (1.0976) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.244, TIME@all 0.305 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:51 loss 1.0783 (1.0819) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0915 (1.0855) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.209, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0831 (1.0754) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0849 (1.0868) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 836.994, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0644 (1.0833) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0696 (1.0941) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.004, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.012) eta 0:44:52 loss 1.0838 (1.0808) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.296 (0.306) data 0.000 (0.006) eta 0:44:44 loss 1.0690 (1.0874) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.057, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0903 (1.0800) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0822 (1.0860) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.026, TIME@all 0.306 -epoch: [175/350][20/50] time 0.306 (0.307) data 0.000 (0.013) eta 0:44:53 loss 1.0688 (1.0774) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.300 (0.307) data 0.000 (0.007) eta 0:44:45 loss 1.0710 (1.0929) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.329, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.014) eta 0:44:52 loss 1.0835 (1.0798) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0606 (1.0902) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.148, TIME@all 0.306 -epoch: [175/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:44:52 loss 1.0902 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:44:44 loss 1.0882 (1.0891) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.204, TIME@all 0.306 -epoch: [176/350][20/50] time 0.315 (0.307) data 0.001 (0.013) eta 0:44:39 loss 1.1476 (1.0873) acc 96.8750 (99.8438) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0879 (1.0960) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.965, TIME@all 0.306 -epoch: [176/350][20/50] time 0.314 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0985 (1.0879) acc 100.0000 (99.5312) lr 0.026000 -epoch: [176/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0990 (1.0960) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.015, TIME@all 0.305 -epoch: [176/350][20/50] time 0.314 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0953 (1.0797) acc 100.0000 (100.0000) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1000 (1.0968) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.972, TIME@all 0.305 -epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0923 (1.0852) acc 100.0000 (99.6875) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1039 (1.0962) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 838.042, TIME@all 0.305 -epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0844 (1.0767) acc 100.0000 (100.0000) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.2082 (1.0915) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 837.971, TIME@all 0.305 -epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.014) eta 0:44:39 loss 1.1010 (1.0848) acc 100.0000 (99.8438) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.0924 (1.0995) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 838.162, TIME@all 0.305 -epoch: [176/350][20/50] time 0.315 (0.307) data 0.000 (0.014) eta 0:44:39 loss 1.1204 (1.0879) acc 100.0000 (99.6875) lr 0.026000 -epoch: [176/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:44:25 loss 1.1311 (1.0962) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.104, TIME@all 0.305 -epoch: [176/350][20/50] time 0.311 (0.307) data 0.000 (0.013) eta 0:44:39 loss 1.0779 (1.0862) acc 100.0000 (100.0000) lr 0.026000 -epoch: [176/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:44:26 loss 1.1732 (1.0959) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 838.207, TIME@all 0.305 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:44:00 loss 1.0938 (1.0850) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:43:59 loss 1.1017 (1.0953) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.794, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:44:00 loss 1.1266 (1.0863) acc 100.0000 (99.8438) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.1223 (1.0937) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.842, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:43:59 loss 1.1261 (1.0812) acc 100.0000 (99.8438) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.0937 (1.0960) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.991, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:44:00 loss 1.2402 (1.0837) acc 93.7500 (99.5312) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.0905 (1.0990) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 840.785, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.1264 (1.0785) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:43:59 loss 1.1582 (1.0972) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.806, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.0955 (1.0961) acc 100.0000 (99.3750) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.1026 (1.0975) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.945, TIME@all 0.304 -epoch: [177/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:43:59 loss 1.1564 (1.0841) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:59 loss 1.1819 (1.0958) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 840.842, TIME@all 0.304 -epoch: [177/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:44:00 loss 1.1029 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:43:58 loss 1.0826 (1.0867) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.198, TIME@all 0.304 -epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:43:59 loss 1.0843 (1.0897) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0918 (1.1021) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 840.205, TIME@all 0.305 -epoch: [178/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.1168 (1.0855) acc 100.0000 (99.6875) lr 0.026000 -epoch: [178/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0848 (1.0928) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.339, TIME@all 0.305 -epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.0849 (1.0871) acc 100.0000 (100.0000) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0972 (1.0947) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.255, TIME@all 0.305 -epoch: [178/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.0930 (1.0831) acc 100.0000 (100.0000) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0687 (1.0890) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.236, TIME@all 0.305 -epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:43:59 loss 1.0773 (1.0884) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:43:48 loss 1.0994 (1.0999) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.254, TIME@all 0.305 -epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1762 (1.0902) acc 96.8750 (99.5312) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0863 (1.1008) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.416, TIME@all 0.305 -epoch: [178/350][20/50] time 0.306 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1203 (1.0852) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0876 (1.0985) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.575, TIME@all 0.305 -epoch: [178/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:43:59 loss 1.1933 (1.0874) acc 100.0000 (99.6875) lr 0.026000 -epoch: [178/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:48 loss 1.0880 (1.1002) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 840.373, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1234 (1.0828) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1336 (1.0887) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.362, TIME@all 0.305 -epoch: [179/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1101 (1.0748) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1895 (1.0930) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 840.314, TIME@all 0.305 -epoch: [179/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:43:39 loss 1.1265 (1.0793) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.0877 (1.0907) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.381, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.1044 (1.0797) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.0894 (1.0863) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.528, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0925 (1.0756) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.2154 (1.0915) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.320, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0853 (1.0881) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1472 (1.0952) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.342, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0901 (1.0787) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1053 (1.0909) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.480, TIME@all 0.305 -epoch: [179/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:43:39 loss 1.0974 (1.0749) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:43:33 loss 1.1647 (1.0940) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 840.684, TIME@all 0.305 -epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.011) eta 0:43:20 loss 1.1295 (1.0868) acc 100.0000 (99.8438) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.1582 (1.0974) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.437, TIME@all 0.305 -epoch: [180/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:43:14 loss 1.2004 (1.0887) acc 96.8750 (99.8438) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:14 loss 1.0862 (1.1024) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 841.163, TIME@all 0.304 -epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1916 (1.0899) acc 96.8750 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.0823 (1.0969) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.458, TIME@all 0.305 -epoch: [180/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 0:43:18 loss 1.0948 (1.0833) acc 100.0000 (99.5312) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:16 loss 1.1251 (1.0953) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 840.719, TIME@all 0.305 -epoch: [180/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1775 (1.0899) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:16 loss 1.0658 (1.0946) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.600, TIME@all 0.305 -epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:43:20 loss 1.1046 (1.0907) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:43:17 loss 1.0815 (1.0913) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.428, TIME@all 0.305 -epoch: [180/350][20/50] time 0.300 (0.304) data 0.001 (0.012) eta 0:43:17 loss 1.0795 (1.0847) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:43:14 loss 1.1400 (1.0990) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 841.154, TIME@all 0.304 -epoch: [180/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:43:20 loss 1.0982 (1.0860) acc 100.0000 (99.8438) lr 0.026000 -epoch: [180/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:43:16 loss 1.0839 (1.0947) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.669, TIME@all 0.305 -epoch: [181/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:43:02 loss 1.0825 (1.0916) acc 100.0000 (99.6875) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:42:54 loss 1.0889 (1.0992) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 842.329, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:43:01 loss 1.1064 (1.0889) acc 100.0000 (100.0000) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1264 (1.0984) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.433, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:43:01 loss 1.1146 (1.0857) acc 100.0000 (100.0000) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1273 (1.0963) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.552, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:43:02 loss 1.0627 (1.0895) acc 100.0000 (99.8438) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:42:54 loss 1.0977 (1.0925) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.348, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:43:02 loss 1.0741 (1.0887) acc 100.0000 (100.0000) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:42:54 loss 1.0908 (1.0992) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.352, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:43:01 loss 1.0761 (1.0911) acc 100.0000 (99.6875) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1602 (1.0938) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.496, TIME@all 0.304 -epoch: [181/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:43:01 loss 1.1279 (1.0923) acc 100.0000 (99.8438) lr 0.026000 -epoch: [181/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1315 (1.0941) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.373, TIME@all 0.304 -epoch: [181/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:43:02 loss 1.0997 (1.0975) acc 100.0000 (99.8438) lr 0.026000 -epoch: [181/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:42:54 loss 1.1215 (1.0933) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.630, TIME@all 0.304 -epoch: [182/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:42:53 loss 1.1315 (1.0749) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1302 (1.0847) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.382, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:53 loss 1.0820 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1499 (1.0840) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.473, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:53 loss 1.0878 (1.0779) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:44 loss 1.1183 (1.0847) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 840.639, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:54 loss 1.2511 (1.0820) acc 96.8750 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0935 (1.0866) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.392, TIME@all 0.305 -epoch: [182/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:42:54 loss 1.0980 (1.0756) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:45 loss 1.1806 (1.0906) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 840.446, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:54 loss 1.0899 (1.0757) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:42:45 loss 1.1181 (1.0851) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.408, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:53 loss 1.1017 (1.0758) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0994 (1.0880) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.563, TIME@all 0.305 -epoch: [182/350][20/50] time 0.302 (0.305) data 0.001 (0.013) eta 0:42:53 loss 1.1058 (1.0822) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:42:45 loss 1.0831 (1.0855) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.829, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.0898 (1.0774) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0816 (1.0859) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 842.097, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.1141 (1.0894) acc 100.0000 (99.6875) lr 0.026000 -epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.1966 (1.0990) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 842.147, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:35 loss 1.1022 (1.0929) acc 100.0000 (99.8438) lr 0.026000 -epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.1067 (1.0951) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.282, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.0924 (1.0953) acc 100.0000 (99.5312) lr 0.026000 -epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.1118 (1.0986) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.149, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:35 loss 1.0879 (1.0873) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.0987 (1.0960) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 842.320, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:42:36 loss 1.1427 (1.0852) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:42:26 loss 1.1185 (1.0944) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.124, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:35 loss 1.1470 (1.1060) acc 100.0000 (99.6875) lr 0.026000 -epoch: [183/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0990 (1.1068) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.498, TIME@all 0.304 -epoch: [183/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:42:36 loss 1.1909 (1.0989) acc 100.0000 (99.6875) lr 0.026000 -epoch: [183/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:42:26 loss 1.0890 (1.0994) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.115, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:42:20 loss 1.0758 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.006) eta 0:42:15 loss 1.1561 (1.0894) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 841.428, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:42:20 loss 1.0826 (1.0784) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.006) eta 0:42:15 loss 1.1021 (1.0866) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.491, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.1034 (1.0818) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.1377 (1.0914) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 841.460, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0977 (1.0763) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.1115 (1.0902) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.453, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0742 (1.0788) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.0715 (1.0845) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.502, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0855 (1.0769) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:15 loss 1.0815 (1.0865) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.745, TIME@all 0.304 -epoch: [184/350][20/50] time 0.303 (0.305) data 0.001 (0.014) eta 0:42:20 loss 1.0935 (1.0787) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.305) data 0.000 (0.007) eta 0:42:14 loss 1.0914 (1.0867) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.572, TIME@all 0.304 -epoch: [184/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:42:20 loss 1.0689 (1.0774) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.314 (0.305) data 0.000 (0.007) eta 0:42:14 loss 1.1037 (1.0888) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.592, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:42:03 loss 1.1332 (1.0764) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:41:54 loss 1.0604 (1.0881) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 842.940, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:42:03 loss 1.1129 (1.0811) acc 100.0000 (99.6875) lr 0.026000 -epoch: [185/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:41:54 loss 1.0830 (1.0952) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.882, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.0645 (1.0779) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0624 (1.0922) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.944, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.1110 (1.0806) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0781 (1.0926) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.892, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.012) eta 0:42:03 loss 1.1449 (1.0784) acc 96.8750 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:41:54 loss 1.0832 (1.0891) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.894, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.013) eta 0:42:03 loss 1.1921 (1.0819) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:41:53 loss 1.0601 (1.0963) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.112, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:42:03 loss 1.0882 (1.0811) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:41:53 loss 1.0668 (1.0914) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.035, TIME@all 0.304 -epoch: [185/350][20/50] time 0.313 (0.305) data 0.001 (0.013) eta 0:42:03 loss 1.0927 (1.0848) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:41:54 loss 1.0713 (1.0951) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.217, TIME@all 0.304 -epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:41:41 loss 1.1244 (1.0743) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:41:37 loss 1.0641 (1.0859) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.076, TIME@all 0.304 -epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0935 (1.0737) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0593 (1.0848) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.173, TIME@all 0.304 -epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0785 (1.0685) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0663 (1.0835) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.189, TIME@all 0.304 -epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1074 (1.0759) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0680 (1.0844) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.109, TIME@all 0.304 -epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.0837 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0792 (1.0836) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.280, TIME@all 0.304 -epoch: [186/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:41:40 loss 1.1047 (1.0785) acc 100.0000 (99.6875) lr 0.026000 -epoch: [186/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:41:36 loss 1.0707 (1.0880) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.334, TIME@all 0.304 -epoch: [186/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1067 (1.0828) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:41:37 loss 1.0587 (1.0880) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.139, TIME@all 0.304 -epoch: [186/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:41:40 loss 1.1064 (1.0738) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.300 (0.304) data 0.001 (0.007) eta 0:41:36 loss 1.0659 (1.0814) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.617, TIME@all 0.303 -epoch: [187/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:41:39 loss 1.1358 (1.0806) acc 96.8750 (99.8438) lr 0.026000 -epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:41:30 loss 1.0969 (1.0876) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.261, TIME@all 0.305 -epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1148 (1.0778) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1866 (1.0903) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.313, TIME@all 0.305 -epoch: [187/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1139 (1.0818) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1103 (1.0906) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.339, TIME@all 0.305 -epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.1115 (1.0802) acc 96.8750 (99.6875) lr 0.026000 -epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:30 loss 1.2085 (1.0901) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.302, TIME@all 0.305 -epoch: [187/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:41:39 loss 1.1633 (1.0908) acc 96.8750 (99.8438) lr 0.026000 -epoch: [187/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:41:30 loss 1.1529 (1.0916) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.277, TIME@all 0.305 -epoch: [187/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:41:38 loss 1.1273 (1.0851) acc 100.0000 (99.8438) lr 0.026000 -epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.2969 (1.0920) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 840.464, TIME@all 0.305 -epoch: [187/350][20/50] time 0.303 (0.305) data 0.001 (0.014) eta 0:41:38 loss 1.0919 (1.0798) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.1721 (1.0895) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 840.414, TIME@all 0.305 -epoch: [187/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:41:38 loss 1.0984 (1.0829) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:41:29 loss 1.0884 (1.0881) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.634, TIME@all 0.305 -epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:15 loss 1.1137 (1.0771) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.0762 (1.0858) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.684, TIME@all 0.305 -epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:15 loss 1.0816 (1.0748) acc 100.0000 (99.8438) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.1035 (1.0870) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.730, TIME@all 0.304 -epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:16 loss 1.0879 (1.0708) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:41:14 loss 1.0671 (1.0837) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.629, TIME@all 0.305 -epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:16 loss 1.1108 (1.0768) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.1025 (1.0843) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.649, TIME@all 0.305 -epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:16 loss 1.0714 (1.0757) acc 100.0000 (99.8438) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0663 (1.0889) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.658, TIME@all 0.305 -epoch: [188/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:41:15 loss 1.0766 (1.0687) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0740 (1.0817) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.855, TIME@all 0.304 -epoch: [188/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:41:15 loss 1.0751 (1.0830) acc 100.0000 (99.8438) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:41:14 loss 1.1053 (1.0964) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.980, TIME@all 0.304 -epoch: [188/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:41:15 loss 1.1182 (1.0757) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:41:14 loss 1.0728 (1.0902) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.802, TIME@all 0.304 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.1911 (1.0755) acc 96.8750 (99.6875) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.0663 (1.0898) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.173, TIME@all 0.304 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.0824 (1.0798) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:40:49 loss 1.1035 (1.0925) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.227, TIME@all 0.304 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:40:58 loss 1.0920 (1.0772) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.0680 (1.0873) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.188, TIME@all 0.304 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.0939 (1.0768) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:50 loss 1.0706 (1.0869) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.145, TIME@all 0.304 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.1769 (1.0824) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:40:49 loss 1.0679 (1.0904) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.348, TIME@all 0.304 -epoch: [189/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:40:57 loss 1.1296 (1.0735) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:50 loss 1.1038 (1.0838) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 843.520, TIME@all 0.303 -epoch: [189/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:40:58 loss 1.1004 (1.0691) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:49 loss 1.0810 (1.0910) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.321, TIME@all 0.304 -epoch: [189/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:40:58 loss 1.1205 (1.0781) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:40:50 loss 1.0726 (1.0976) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.155, TIME@all 0.304 -epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0997 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.1232 (1.0870) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.194, TIME@all 0.304 -epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0759 (1.0741) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.1019 (1.0856) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.265, TIME@all 0.304 -epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0521 (1.0749) acc 100.0000 (99.6875) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:41 loss 1.1127 (1.0861) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.221, TIME@all 0.304 -epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0593 (1.0714) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:41 loss 1.1249 (1.0884) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.194, TIME@all 0.304 -epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0689 (1.0694) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:40 loss 1.1196 (1.0801) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 842.396, TIME@all 0.304 -epoch: [190/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:40:42 loss 1.0892 (1.0739) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:40:40 loss 1.0838 (1.0822) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 842.330, TIME@all 0.304 -epoch: [190/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0738 (1.0747) acc 100.0000 (99.8438) lr 0.026000 -epoch: [190/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.0921 (1.0826) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.023, TIME@all 0.304 -epoch: [190/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:40:42 loss 1.0892 (1.0713) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:40:41 loss 1.0775 (1.0849) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.489, TIME@all 0.304 -epoch: [191/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:22 loss 1.0762 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1390 (1.0815) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.239, TIME@all 0.304 -epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.2040 (1.0854) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1337 (1.0900) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.271, TIME@all 0.304 -epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.1190 (1.0815) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1556 (1.0932) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 843.305, TIME@all 0.304 -epoch: [191/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:40:21 loss 1.0775 (1.0721) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1615 (1.0819) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.265, TIME@all 0.304 -epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.1098 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.0990 (1.0889) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.271, TIME@all 0.304 -epoch: [191/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:40:21 loss 1.0822 (1.0749) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:40:19 loss 1.1616 (1.0908) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.452, TIME@all 0.304 -epoch: [191/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:40:21 loss 1.0906 (1.0733) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:40:19 loss 1.1432 (1.0813) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.417, TIME@all 0.304 -epoch: [191/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:40:21 loss 1.0961 (1.0755) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:40:20 loss 1.1551 (1.0904) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 843.560, TIME@all 0.303 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:13 loss 1.0720 (1.0775) acc 100.0000 (99.6875) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.006) eta 0:40:07 loss 1.0744 (1.0804) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.929, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:12 loss 1.0866 (1.0703) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 0:40:07 loss 1.0975 (1.0788) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.961, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:13 loss 1.0647 (1.0708) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.1359 (1.0900) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.938, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:12 loss 1.0649 (1.0633) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.0893 (1.0813) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.154, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:13 loss 1.1015 (1.0683) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.2211 (1.0875) acc 93.7500 (99.7656) lr 0.026000 -FPS@all 841.954, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:40:12 loss 1.0979 (1.0701) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.1018 (1.0821) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.367, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.001 (0.014) eta 0:40:12 loss 1.0852 (1.0719) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.296 (0.304) data 0.000 (0.007) eta 0:40:07 loss 1.0892 (1.0889) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.137, TIME@all 0.304 -epoch: [192/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:40:12 loss 1.0676 (1.0685) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:40:08 loss 1.1376 (1.0890) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.974, TIME@all 0.304 -epoch: [193/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0706 (1.0697) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1734 (1.0817) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 840.206, TIME@all 0.305 -epoch: [193/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:40:10 loss 1.0917 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1410 (1.0838) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 840.306, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:10 loss 1.0568 (1.0699) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.001 (0.007) eta 0:39:57 loss 1.0762 (1.0810) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.424, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0924 (1.0750) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.0833 (1.0757) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.219, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:11 loss 1.0766 (1.0805) acc 100.0000 (99.8438) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:39:58 loss 1.1147 (1.0863) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.243, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.013) eta 0:40:10 loss 1.0828 (1.0716) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:39:58 loss 1.1360 (1.0827) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 840.342, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:11 loss 1.0765 (1.0743) acc 100.0000 (99.8438) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1021 (1.0830) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.212, TIME@all 0.305 -epoch: [193/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:40:10 loss 1.1034 (1.0792) acc 96.8750 (99.8438) lr 0.026000 -epoch: [193/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:39:58 loss 1.1485 (1.0850) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.512, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:39:54 loss 1.0764 (1.0712) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:44 loss 1.0859 (1.0786) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.086, TIME@all 0.305 -epoch: [194/350][20/50] time 0.306 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0769 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.1518 (1.0819) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.115, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.011) eta 0:39:54 loss 1.0862 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0691 (1.0857) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.149, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.001 (0.013) eta 0:39:53 loss 1.1327 (1.0784) acc 100.0000 (99.8438) lr 0.026000 -epoch: [194/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:39:43 loss 1.0631 (1.0872) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.290, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.1074 (1.0802) acc 100.0000 (99.8438) lr 0.026000 -epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0865 (1.0831) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.085, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0850 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.1142 (1.0794) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.091, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:53 loss 1.1072 (1.0743) acc 96.8750 (99.8438) lr 0.026000 -epoch: [194/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0828 (1.0843) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.245, TIME@all 0.305 -epoch: [194/350][20/50] time 0.305 (0.306) data 0.000 (0.012) eta 0:39:54 loss 1.0740 (1.0744) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:39:43 loss 1.0729 (1.0821) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.411, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:39:34 loss 1.0978 (1.0745) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:39:26 loss 1.1371 (1.0859) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 840.469, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.0768 (1.0716) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.0714 (1.0792) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.510, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1576 (1.0793) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1347 (1.0925) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.493, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1330 (1.0726) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1062 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.650, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.0784 (1.0746) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1443 (1.0883) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 840.482, TIME@all 0.305 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1254 (1.0777) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1343 (1.0875) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.692, TIME@all 0.305 -epoch: [195/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1309 (1.0802) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1336 (1.0955) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.858, TIME@all 0.304 -epoch: [195/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:39:34 loss 1.1697 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:39:26 loss 1.1066 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.521, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.0713 (1.0798) acc 100.0000 (99.8438) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.306) data 0.000 (0.006) eta 0:39:15 loss 1.1081 (1.0856) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.022, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.1024 (1.0722) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.1029 (1.0882) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.043, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.1102 (1.0710) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.0579 (1.0860) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.065, TIME@all 0.305 -epoch: [196/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:39:18 loss 1.0712 (1.0742) acc 100.0000 (99.8438) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.006) eta 0:39:15 loss 1.1236 (1.0858) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.029, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:17 loss 1.1019 (1.0776) acc 100.0000 (99.8438) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.0874 (1.0841) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.162, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:17 loss 1.0673 (1.0733) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.1012 (1.0816) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.209, TIME@all 0.305 -epoch: [196/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:39:18 loss 1.0744 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.323 (0.305) data 0.000 (0.007) eta 0:39:14 loss 1.0970 (1.0854) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.385, TIME@all 0.305 -epoch: [196/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:39:18 loss 1.1490 (1.0835) acc 96.8750 (99.3750) lr 0.026000 -epoch: [196/350][40/50] time 0.327 (0.305) data 0.000 (0.007) eta 0:39:15 loss 1.1000 (1.0923) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.004, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0633 (1.0776) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0923 (1.0889) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.789, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0898 (1.0820) acc 100.0000 (99.6875) lr 0.026000 -epoch: [197/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.1071 (1.0949) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 839.891, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:10 loss 1.0657 (1.0700) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0859 (1.0967) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 839.767, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.012) eta 0:39:09 loss 1.0676 (1.0760) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:57 loss 1.0819 (1.0881) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.795, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:10 loss 1.0701 (1.0740) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:57 loss 1.1597 (1.0882) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 839.814, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:09 loss 1.0605 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.0773 (1.0890) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.967, TIME@all 0.305 -epoch: [197/350][20/50] time 0.319 (0.306) data 0.000 (0.013) eta 0:39:09 loss 1.0583 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.0884 (1.0827) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.935, TIME@all 0.305 -epoch: [197/350][20/50] time 0.322 (0.306) data 0.000 (0.013) eta 0:39:11 loss 1.0565 (1.0743) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:38:56 loss 1.1044 (1.0906) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.139, TIME@all 0.305 -epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:47 loss 1.0775 (1.0699) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:38:44 loss 1.0753 (1.0768) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.242, TIME@all 0.305 -epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:38:47 loss 1.0640 (1.0700) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0673 (1.0808) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.399, TIME@all 0.305 -epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0825 (1.0680) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:44 loss 1.0575 (1.0777) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.252, TIME@all 0.305 -epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0772 (1.0720) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0818 (1.0814) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.274, TIME@all 0.305 -epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:38:47 loss 1.1205 (1.0818) acc 100.0000 (99.6875) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0847 (1.0880) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.462, TIME@all 0.305 -epoch: [198/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0718 (1.0673) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:38:44 loss 1.0758 (1.0868) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.251, TIME@all 0.305 -epoch: [198/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.1091 (1.0720) acc 100.0000 (99.6875) lr 0.026000 -epoch: [198/350][40/50] time 0.299 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0737 (1.0781) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.583, TIME@all 0.305 -epoch: [198/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:47 loss 1.0772 (1.0739) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:38:43 loss 1.0923 (1.0811) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.335, TIME@all 0.305 -epoch: [199/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1300 (1.0689) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:26 loss 1.0975 (1.0821) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.821, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1368 (1.0733) acc 100.0000 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:26 loss 1.0894 (1.0849) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.835, TIME@all 0.305 -epoch: [199/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:38:28 loss 1.1427 (1.0710) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:38:25 loss 1.1172 (1.0884) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.926, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.1263 (1.0702) acc 100.0000 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1187 (1.0849) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.841, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.2185 (1.0727) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1371 (1.0894) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.864, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:38:27 loss 1.2002 (1.0787) acc 96.8750 (99.6875) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:38:25 loss 1.1862 (1.0901) acc 96.8750 (99.4531) lr 0.026000 -FPS@all 840.042, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:38:28 loss 1.1273 (1.0732) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:25 loss 1.1778 (1.0901) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 839.981, TIME@all 0.305 -epoch: [199/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:38:28 loss 1.2240 (1.0768) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:38:26 loss 1.1324 (1.0835) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.157, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:38:19 loss 1.0734 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.1385 (1.0881) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.753, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:18 loss 1.1050 (1.0825) acc 100.0000 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.1325 (1.0899) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 838.788, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:19 loss 1.1591 (1.0749) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.006) eta 0:38:16 loss 1.0606 (1.0908) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.807, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:19 loss 1.1619 (1.0907) acc 96.8750 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:38:16 loss 1.1029 (1.0899) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.747, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:19 loss 1.1258 (1.0781) acc 100.0000 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.306 (0.306) data 0.001 (0.006) eta 0:38:16 loss 1.0704 (1.0875) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.789, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:18 loss 1.1974 (1.0820) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.306 (0.306) data 0.001 (0.007) eta 0:38:16 loss 1.0995 (1.0961) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.955, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:38:18 loss 1.0835 (1.0804) acc 100.0000 (99.6875) lr 0.026000 -epoch: [200/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:38:16 loss 1.1183 (1.0909) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.899, TIME@all 0.305 -epoch: [200/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:38:17 loss 1.0746 (1.0856) acc 100.0000 (99.6875) lr 0.026000 -epoch: [200/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:38:15 loss 1.1502 (1.0932) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 839.256, TIME@all 0.305 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1089 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0668 (1.0827) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.159, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.0999 (1.0773) acc 100.0000 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:52 loss 1.0658 (1.0896) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 841.194, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.0921 (1.0838) acc 100.0000 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0735 (1.0893) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.155, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.014) eta 0:38:01 loss 1.0928 (1.0802) acc 100.0000 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.001 (0.007) eta 0:37:52 loss 1.0699 (1.0909) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 841.355, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1514 (1.0760) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0702 (1.0856) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.175, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:02 loss 1.1548 (1.0805) acc 96.8750 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:37:53 loss 1.0718 (1.0922) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.504, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:38:02 loss 1.1179 (1.0730) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:37:52 loss 1.0854 (1.0886) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.222, TIME@all 0.304 -epoch: [201/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:38:01 loss 1.0994 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:37:52 loss 1.0724 (1.0899) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.339, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0810 (1.0721) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0675 (1.0819) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.431, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0757 (1.0792) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0879 (1.0827) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.457, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.1610 (1.0775) acc 96.8750 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0855 (1.0839) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.421, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:37:49 loss 1.0777 (1.0735) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:37:38 loss 1.0886 (1.0799) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.627, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.001 (0.012) eta 0:37:49 loss 1.0914 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0788 (1.0790) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.428, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0826 (1.0705) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.1080 (1.0833) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.413, TIME@all 0.304 -epoch: [202/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:37:49 loss 1.0607 (1.0799) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:37:38 loss 1.0655 (1.0829) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.580, TIME@all 0.304 -epoch: [202/350][20/50] time 0.308 (0.305) data 0.000 (0.012) eta 0:37:49 loss 1.0817 (1.0771) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:37:39 loss 1.0676 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.744, TIME@all 0.304 -epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.1311 (1.0797) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.1744 (1.0866) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 840.996, TIME@all 0.304 -epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.0978 (1.0752) acc 100.0000 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.0951 (1.0830) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.034, TIME@all 0.304 -epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:37:33 loss 1.0837 (1.0715) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.1510 (1.0848) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 841.041, TIME@all 0.304 -epoch: [203/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:37:34 loss 1.0989 (1.0769) acc 100.0000 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.007) eta 0:37:24 loss 1.1125 (1.0870) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 841.004, TIME@all 0.304 -epoch: [203/350][20/50] time 0.303 (0.305) data 0.001 (0.012) eta 0:37:33 loss 1.1456 (1.0753) acc 96.8750 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.006) eta 0:37:24 loss 1.3520 (1.0909) acc 93.7500 (99.7656) lr 0.026000 -FPS@all 841.021, TIME@all 0.304 -epoch: [203/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.1581 (1.0795) acc 96.8750 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.000 (0.007) eta 0:37:23 loss 1.1120 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.193, TIME@all 0.304 -epoch: [203/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.1652 (1.0826) acc 96.8750 (99.5312) lr 0.026000 -epoch: [203/350][40/50] time 0.309 (0.305) data 0.001 (0.007) eta 0:37:23 loss 1.0745 (1.0854) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.367, TIME@all 0.304 -epoch: [203/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:37:33 loss 1.0812 (1.0651) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.312 (0.305) data 0.001 (0.007) eta 0:37:23 loss 1.0895 (1.0762) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.104, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.011) eta 0:37:18 loss 1.0770 (1.0723) acc 96.8750 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0638 (1.0853) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.645, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0595 (1.0647) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0606 (1.0802) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.678, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0769 (1.0748) acc 100.0000 (99.6875) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0619 (1.0853) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.736, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0683 (1.0737) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0533 (1.0885) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.649, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0613 (1.0757) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0857 (1.0865) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.680, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.013) eta 0:37:18 loss 1.0793 (1.0810) acc 100.0000 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.007) eta 0:37:06 loss 1.0618 (1.0916) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.873, TIME@all 0.304 -epoch: [204/350][20/50] time 0.309 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0660 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.316 (0.305) data 0.000 (0.006) eta 0:37:08 loss 1.0600 (1.0848) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.967, TIME@all 0.304 -epoch: [204/350][20/50] time 0.313 (0.305) data 0.000 (0.012) eta 0:37:18 loss 1.0689 (1.0670) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.311 (0.305) data 0.000 (0.006) eta 0:37:07 loss 1.0680 (1.0834) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.814, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.012) eta 0:37:01 loss 1.0843 (1.0681) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0768 (1.0763) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.539, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.012) eta 0:37:01 loss 1.0716 (1.0675) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0911 (1.0749) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.589, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.013) eta 0:37:01 loss 1.1152 (1.0709) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.0665 (1.0806) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.584, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.001 (0.013) eta 0:37:01 loss 1.1040 (1.0668) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.0618 (1.0863) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 842.770, TIME@all 0.304 -epoch: [205/350][20/50] time 0.313 (0.305) data 0.001 (0.012) eta 0:37:01 loss 1.0708 (1.0630) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0869 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.594, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.001 (0.012) eta 0:37:01 loss 1.0844 (1.0669) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0615 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 842.603, TIME@all 0.304 -epoch: [205/350][20/50] time 0.316 (0.305) data 0.001 (0.012) eta 0:37:02 loss 1.0904 (1.0762) acc 100.0000 (99.8438) lr 0.026000 -epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:49 loss 1.0574 (1.0794) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.895, TIME@all 0.304 -epoch: [205/350][20/50] time 0.314 (0.305) data 0.000 (0.013) eta 0:37:01 loss 1.0719 (1.0704) acc 100.0000 (99.8438) lr 0.026000 -epoch: [205/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:49 loss 1.1234 (1.0830) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 842.734, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0779 (1.0742) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.0735 (1.0815) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.425, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:36:35 loss 1.0721 (1.0738) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.1004 (1.0838) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.345, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0720 (1.0703) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0848 (1.0765) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.363, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0849 (1.0699) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0984 (1.0791) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.361, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:36:34 loss 1.0547 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:35 loss 1.0817 (1.0819) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.548, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0646 (1.0640) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:36:35 loss 1.0709 (1.0737) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 841.362, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:36:35 loss 1.0917 (1.0807) acc 100.0000 (99.6875) lr 0.026000 -epoch: [206/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:36:35 loss 1.0642 (1.0870) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 841.716, TIME@all 0.304 -epoch: [206/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:36:34 loss 1.0692 (1.0694) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:36:35 loss 1.1298 (1.0776) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.504, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:26 loss 1.1383 (1.0760) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.0785 (1.0965) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 841.572, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.1532 (1.0702) acc 96.8750 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:22 loss 1.0614 (1.0930) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 841.585, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.0747 (1.0687) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:22 loss 1.1170 (1.0811) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.561, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:25 loss 1.1071 (1.0645) acc 96.8750 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.1403 (1.0804) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.609, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:36:26 loss 1.1218 (1.0722) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.305 (0.305) data 0.001 (0.006) eta 0:36:22 loss 1.0831 (1.0869) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 841.602, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.1864 (1.0774) acc 96.8750 (99.6875) lr 0.026000 -epoch: [207/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:36:21 loss 1.1160 (1.0937) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 841.783, TIME@all 0.304 -epoch: [207/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:36:25 loss 1.0845 (1.0708) acc 100.0000 (100.0000) lr 0.026000 -epoch: [207/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:36:21 loss 1.0914 (1.0841) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 841.728, TIME@all 0.304 -epoch: [207/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:36:26 loss 1.1531 (1.0726) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:36:22 loss 1.0911 (1.0793) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 841.877, TIME@all 0.304 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.012) eta 0:36:07 loss 1.0852 (1.0809) acc 100.0000 (99.8438) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:36:00 loss 1.0885 (1.0904) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.338, TIME@all 0.304 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:06 loss 1.0797 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0704 (1.0846) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.472, TIME@all 0.304 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0919 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:01 loss 1.0674 (1.0820) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 843.321, TIME@all 0.304 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0528 (1.0764) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0624 (1.0827) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 843.554, TIME@all 0.303 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:06 loss 1.0692 (1.0705) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:36:00 loss 1.1003 (1.0776) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 843.558, TIME@all 0.303 -epoch: [208/350][20/50] time 0.299 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0681 (1.0750) acc 100.0000 (99.8438) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0650 (1.0836) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.367, TIME@all 0.304 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0576 (1.0702) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0680 (1.0846) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.643, TIME@all 0.303 -epoch: [208/350][20/50] time 0.298 (0.304) data 0.000 (0.013) eta 0:36:07 loss 1.0774 (1.0830) acc 100.0000 (99.6875) lr 0.026000 -epoch: [208/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:36:00 loss 1.0705 (1.0834) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.501, TIME@all 0.303 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.0956 (1.0710) acc 100.0000 (99.6875) lr 0.026000 -epoch: [209/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.1262 (1.0838) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 843.046, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.1344 (1.0715) acc 100.0000 (99.8438) lr 0.026000 -epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.0757 (1.0858) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 843.007, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.0894 (1.0663) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:48 loss 1.0998 (1.0803) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 842.952, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:51 loss 1.0638 (1.0679) acc 100.0000 (99.8438) lr 0.026000 -epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:48 loss 1.0755 (1.0773) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.959, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:35:51 loss 1.1042 (1.0776) acc 100.0000 (99.6875) lr 0.026000 -epoch: [209/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:47 loss 1.0753 (1.0924) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 843.159, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:51 loss 1.0639 (1.0681) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:47 loss 1.0722 (1.0878) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 842.989, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:51 loss 1.1466 (1.0741) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:35:47 loss 1.0837 (1.0814) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 843.110, TIME@all 0.304 -epoch: [209/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:50 loss 1.1132 (1.0712) acc 100.0000 (99.6875) lr 0.026000 -epoch: [209/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:35:47 loss 1.0590 (1.0846) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 843.382, TIME@all 0.304 -epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:39 loss 1.0783 (1.0746) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0873 (1.0837) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.391, TIME@all 0.304 -epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0534 (1.0677) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0882 (1.0772) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.428, TIME@all 0.304 -epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0638 (1.0709) acc 100.0000 (99.8438) lr 0.026000 -epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0805 (1.0791) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.427, TIME@all 0.304 -epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0520 (1.0668) acc 100.0000 (99.6875) lr 0.026000 -epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0593 (1.0751) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.426, TIME@all 0.304 -epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.1053 (1.0734) acc 100.0000 (99.8438) lr 0.026000 -epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0858 (1.0854) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 842.647, TIME@all 0.304 -epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0789 (1.0687) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:35:34 loss 1.0701 (1.0798) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.436, TIME@all 0.304 -epoch: [210/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:35:38 loss 1.0534 (1.0647) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:35:33 loss 1.0868 (1.0777) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.581, TIME@all 0.304 -epoch: [210/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:35:38 loss 1.0650 (1.0664) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:35:33 loss 1.1312 (1.0791) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 842.822, TIME@all 0.304 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:35:20 loss 1.0823 (1.0690) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0730 (1.0839) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.911, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:35:21 loss 1.0721 (1.0703) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0679 (1.0824) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.860, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:35:21 loss 1.0702 (1.0699) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0627 (1.0802) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.886, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:35:20 loss 1.0952 (1.0726) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.300 (0.305) data 0.001 (0.007) eta 0:35:23 loss 1.0617 (1.0800) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.103, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:20 loss 1.0783 (1.0708) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:35:23 loss 1.0890 (1.0842) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.896, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:20 loss 1.0583 (1.0716) acc 100.0000 (99.8438) lr 0.026000 -epoch: [211/350][40/50] time 0.300 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0533 (1.0804) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 840.035, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:35:21 loss 1.0793 (1.0705) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0572 (1.0754) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.861, TIME@all 0.305 -epoch: [211/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:35:21 loss 1.0672 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:35:23 loss 1.0575 (1.0827) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.160, TIME@all 0.305 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:23 loss 1.0662 (1.0674) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1529 (1.0917) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 836.194, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:23 loss 1.0753 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1046 (1.0876) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 836.224, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.012) eta 0:35:22 loss 1.0621 (1.0694) acc 100.0000 (99.6875) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1263 (1.0849) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 836.267, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0755 (1.0736) acc 100.0000 (99.8438) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.2314 (1.0939) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 836.186, TIME@all 0.306 -epoch: [212/350][20/50] time 0.304 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0907 (1.0772) acc 100.0000 (99.6875) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:35:17 loss 1.1121 (1.0973) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 836.211, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:22 loss 1.1018 (1.0694) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.1381 (1.0939) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 836.384, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0876 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.0771 (1.0894) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 836.341, TIME@all 0.306 -epoch: [212/350][20/50] time 0.303 (0.306) data 0.000 (0.013) eta 0:35:23 loss 1.0554 (1.0646) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:35:17 loss 1.1238 (1.0776) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 836.530, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.0775 (1.0845) acc 100.0000 (99.6875) lr 0.026000 -epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0737 (1.0908) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.563, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.0834 (1.0644) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0641 (1.0766) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.483, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:35:12 loss 1.1027 (1.0762) acc 100.0000 (99.6875) lr 0.026000 -epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.006) eta 0:35:04 loss 1.0733 (1.0816) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 837.549, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0908 (1.0768) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.327 (0.307) data 0.000 (0.007) eta 0:35:04 loss 1.0677 (1.0817) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.531, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0961 (1.0704) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.326 (0.307) data 0.001 (0.007) eta 0:35:04 loss 1.1024 (1.0781) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.550, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.0830 (1.0681) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.326 (0.307) data 0.000 (0.007) eta 0:35:03 loss 1.0980 (1.0827) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.715, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:12 loss 1.1029 (1.0784) acc 100.0000 (99.8438) lr 0.026000 -epoch: [213/350][40/50] time 0.326 (0.307) data 0.001 (0.007) eta 0:35:03 loss 1.1014 (1.0860) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.670, TIME@all 0.306 -epoch: [213/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:35:11 loss 1.0579 (1.0835) acc 100.0000 (99.5312) lr 0.026000 -epoch: [213/350][40/50] time 0.329 (0.307) data 0.000 (0.007) eta 0:35:03 loss 1.0611 (1.0854) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.941, TIME@all 0.306 -epoch: [214/350][20/50] time 0.324 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.2013 (1.0775) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0886 (1.0915) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.742, TIME@all 0.305 -epoch: [214/350][20/50] time 0.322 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.1080 (1.0786) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0918 (1.0845) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.756, TIME@all 0.305 -epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.011) eta 0:34:47 loss 1.1230 (1.0714) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1352 (1.0889) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.663, TIME@all 0.305 -epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.0961 (1.0806) acc 100.0000 (99.5312) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.0745 (1.0928) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 838.725, TIME@all 0.305 -epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.012) eta 0:34:47 loss 1.0786 (1.0743) acc 100.0000 (99.6875) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1245 (1.0932) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 838.713, TIME@all 0.305 -epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.013) eta 0:34:46 loss 1.1067 (1.0745) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:34:38 loss 1.1209 (1.0901) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.901, TIME@all 0.305 -epoch: [214/350][20/50] time 0.323 (0.306) data 0.000 (0.013) eta 0:34:47 loss 1.1219 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:34:38 loss 1.1124 (1.1006) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.853, TIME@all 0.305 -epoch: [214/350][20/50] time 0.325 (0.306) data 0.000 (0.012) eta 0:34:48 loss 1.0665 (1.0772) acc 100.0000 (99.8438) lr 0.026000 -epoch: [214/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:34:39 loss 1.1498 (1.0960) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 838.999, TIME@all 0.305 -epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:40 loss 1.1533 (1.0800) acc 100.0000 (99.8438) lr 0.026000 -epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0859 (1.0843) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 834.900, TIME@all 0.307 -epoch: [215/350][20/50] time 0.312 (0.307) data 0.000 (0.011) eta 0:34:41 loss 1.0969 (1.0779) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0723 (1.0853) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 834.816, TIME@all 0.307 -epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.1385 (1.0822) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0696 (1.0862) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 834.884, TIME@all 0.307 -epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.1588 (1.0862) acc 100.0000 (99.8438) lr 0.026000 -epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0734 (1.0869) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 834.838, TIME@all 0.307 -epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:34:41 loss 1.0705 (1.0776) acc 100.0000 (99.6875) lr 0.026000 -epoch: [215/350][40/50] time 0.304 (0.307) data 0.000 (0.006) eta 0:34:36 loss 1.0732 (1.0860) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 834.823, TIME@all 0.307 -epoch: [215/350][20/50] time 0.312 (0.307) data 0.000 (0.013) eta 0:34:40 loss 1.0649 (1.0795) acc 100.0000 (99.8438) lr 0.026000 -epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.007) eta 0:34:36 loss 1.1552 (1.0866) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 834.978, TIME@all 0.307 -epoch: [215/350][20/50] time 0.311 (0.307) data 0.000 (0.013) eta 0:34:41 loss 1.1062 (1.0731) acc 100.0000 (99.8438) lr 0.026000 -epoch: [215/350][40/50] time 0.305 (0.307) data 0.000 (0.007) eta 0:34:36 loss 1.0801 (1.0877) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 834.948, TIME@all 0.307 -epoch: [215/350][20/50] time 0.310 (0.307) data 0.000 (0.012) eta 0:34:40 loss 1.0651 (1.0731) acc 100.0000 (99.6875) lr 0.026000 -epoch: [215/350][40/50] time 0.305 (0.307) data 0.001 (0.006) eta 0:34:36 loss 1.0648 (1.0854) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 835.187, TIME@all 0.307 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:34:13 loss 1.1136 (1.0739) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1388 (1.0858) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.563, TIME@all 0.305 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0647 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:34:07 loss 1.0818 (1.0840) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.562, TIME@all 0.305 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1195 (1.0707) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1088 (1.0862) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 840.543, TIME@all 0.305 -epoch: [216/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:34:13 loss 1.1571 (1.0782) acc 100.0000 (99.8438) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.0761 (1.0955) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 840.550, TIME@all 0.305 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1363 (1.0715) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:34:07 loss 1.1599 (1.0844) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 840.574, TIME@all 0.305 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0863 (1.0630) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:34:06 loss 1.0819 (1.0869) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 840.757, TIME@all 0.304 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:34:12 loss 1.0975 (1.0687) acc 100.0000 (99.8438) lr 0.026000 -epoch: [216/350][40/50] time 0.303 (0.305) data 0.000 (0.007) eta 0:34:07 loss 1.1633 (1.0872) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 840.699, TIME@all 0.305 -epoch: [216/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:34:12 loss 1.1000 (1.0728) acc 100.0000 (99.8438) lr 0.026000 -epoch: [216/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:34:06 loss 1.1176 (1.0796) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 840.856, TIME@all 0.304 -epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:34:01 loss 1.1956 (1.0761) acc 96.8750 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:33:57 loss 1.1372 (1.0854) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.836, TIME@all 0.306 -epoch: [217/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:34:00 loss 1.1324 (1.0792) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1301 (1.0891) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.922, TIME@all 0.306 -epoch: [217/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.1098 (1.0721) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.2259 (1.0918) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 837.871, TIME@all 0.306 -epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.0762 (1.0755) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.0816 (1.0908) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.874, TIME@all 0.306 -epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:01 loss 1.1653 (1.0855) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1260 (1.0931) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.884, TIME@all 0.306 -epoch: [217/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:34:00 loss 1.1270 (1.0772) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:57 loss 1.1082 (1.0894) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.057, TIME@all 0.305 -epoch: [217/350][20/50] time 0.302 (0.306) data 0.000 (0.013) eta 0:34:00 loss 1.1038 (1.0757) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:57 loss 1.1269 (1.0803) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.039, TIME@all 0.305 -epoch: [217/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:34:00 loss 1.0955 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [217/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:57 loss 1.0982 (1.0858) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.313, TIME@all 0.305 -epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:33:44 loss 1.0625 (1.0712) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:33:45 loss 1.0750 (1.0790) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.088, TIME@all 0.305 -epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0515 (1.0677) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0694 (1.0821) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.070, TIME@all 0.305 -epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:33:44 loss 1.0642 (1.0718) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:33:45 loss 1.0774 (1.0805) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.162, TIME@all 0.305 -epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0675 (1.0685) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0506 (1.0752) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.094, TIME@all 0.305 -epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0937 (1.0715) acc 96.8750 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:45 loss 1.0729 (1.0771) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.096, TIME@all 0.305 -epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0630 (1.0704) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:33:44 loss 1.1334 (1.0859) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.277, TIME@all 0.305 -epoch: [218/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:33:44 loss 1.0969 (1.0693) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:33:44 loss 1.0562 (1.0813) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.214, TIME@all 0.305 -epoch: [218/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:33:43 loss 1.0796 (1.0717) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:33:44 loss 1.0829 (1.0868) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 838.494, TIME@all 0.305 -epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:33:41 loss 1.0757 (1.0658) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:28 loss 1.0570 (1.0759) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.622, TIME@all 0.306 -epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0672 (1.0745) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:28 loss 1.0857 (1.0869) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.649, TIME@all 0.306 -epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:33:41 loss 1.0549 (1.0699) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:33:27 loss 1.0654 (1.0803) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.690, TIME@all 0.306 -epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0769 (1.0727) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:28 loss 1.0623 (1.0850) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.636, TIME@all 0.306 -epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:40 loss 1.0606 (1.0645) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:27 loss 1.0706 (1.0827) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.827, TIME@all 0.306 -epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0695 (1.0670) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:33:27 loss 1.0768 (1.0848) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.769, TIME@all 0.306 -epoch: [219/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0857 (1.0665) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.301 (0.306) data 0.001 (0.007) eta 0:33:28 loss 1.0589 (1.0837) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.629, TIME@all 0.306 -epoch: [219/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:33:41 loss 1.0602 (1.0683) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:33:27 loss 1.0589 (1.0874) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.987, TIME@all 0.305 -epoch: [220/350][20/50] time 0.311 (0.307) data 0.000 (0.012) eta 0:33:24 loss 1.0758 (1.0767) acc 100.0000 (99.6875) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.006) eta 0:33:15 loss 1.0909 (1.0845) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.885, TIME@all 0.306 -epoch: [220/350][20/50] time 0.309 (0.307) data 0.001 (0.012) eta 0:33:24 loss 1.0693 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.006) eta 0:33:15 loss 1.0731 (1.0821) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.970, TIME@all 0.306 -epoch: [220/350][20/50] time 0.306 (0.307) data 0.001 (0.013) eta 0:33:24 loss 1.0873 (1.0737) acc 96.8750 (99.8438) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.0889 (1.0848) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.880, TIME@all 0.306 -epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0792 (1.0692) acc 100.0000 (99.8438) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.1520 (1.0837) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 837.872, TIME@all 0.306 -epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:23 loss 1.0592 (1.0676) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:14 loss 1.0703 (1.0788) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.110, TIME@all 0.305 -epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0675 (1.0653) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:15 loss 1.0984 (1.0785) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.914, TIME@all 0.306 -epoch: [220/350][20/50] time 0.310 (0.307) data 0.000 (0.013) eta 0:33:24 loss 1.0876 (1.0683) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:33:14 loss 1.0830 (1.0770) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.059, TIME@all 0.305 -epoch: [220/350][20/50] time 0.308 (0.307) data 0.000 (0.013) eta 0:33:23 loss 1.1131 (1.0733) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.316 (0.307) data 0.000 (0.007) eta 0:33:15 loss 1.0848 (1.0822) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 838.322, TIME@all 0.305 -epoch: [221/350][20/50] time 0.309 (0.306) data 0.000 (0.011) eta 0:33:00 loss 1.0975 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0945 (1.0813) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.319, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1292 (1.0724) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.1764 (1.0822) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 837.375, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1014 (1.0801) acc 100.0000 (99.8438) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0786 (1.0777) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.354, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.1455 (1.0768) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.303 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.1198 (1.0803) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.408, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.001 (0.013) eta 0:32:59 loss 1.1493 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:56 loss 1.1349 (1.0793) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.539, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:33:00 loss 1.0901 (1.0794) acc 100.0000 (99.8438) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0819 (1.0790) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 837.368, TIME@all 0.306 -epoch: [221/350][20/50] time 0.316 (0.306) data 0.000 (0.012) eta 0:33:01 loss 1.0670 (1.0719) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.305 (0.306) data 0.001 (0.006) eta 0:32:57 loss 1.0988 (1.0774) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.672, TIME@all 0.306 -epoch: [221/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:32:59 loss 1.0991 (1.0730) acc 100.0000 (99.8438) lr 0.026000 -epoch: [221/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:57 loss 1.0865 (1.0740) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 837.500, TIME@all 0.306 -epoch: [222/350][20/50] time 0.301 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0676 (1.0747) acc 100.0000 (99.8438) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0529 (1.0845) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 838.052, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0531 (1.0679) acc 100.0000 (99.8438) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0866 (1.0780) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 838.096, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.011) eta 0:32:44 loss 1.0899 (1.0648) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0828 (1.0783) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 837.999, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0808 (1.0694) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.1238 (1.0806) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 838.023, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0731 (1.0650) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0555 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 838.040, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:32:44 loss 1.0548 (1.0638) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0711 (1.0715) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 838.225, TIME@all 0.305 -epoch: [222/350][20/50] time 0.304 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0827 (1.0741) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.301 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.0632 (1.0776) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 838.351, TIME@all 0.305 -epoch: [222/350][20/50] time 0.302 (0.306) data 0.000 (0.012) eta 0:32:44 loss 1.0567 (1.0658) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:32:40 loss 1.1484 (1.0834) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 838.169, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.1059 (1.0746) acc 100.0000 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.308 (0.306) data 0.001 (0.006) eta 0:32:23 loss 1.1427 (1.0850) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.078, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.1014 (1.0643) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1549 (1.0771) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.116, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.0675 (1.0661) acc 100.0000 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.309 (0.306) data 0.001 (0.006) eta 0:32:23 loss 1.1534 (1.0822) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.130, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0742 (1.0679) acc 100.0000 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:32:23 loss 1.0902 (1.0750) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.297, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.012) eta 0:32:24 loss 1.0806 (1.0624) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.308 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1209 (1.0789) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.098, TIME@all 0.305 -epoch: [223/350][20/50] time 0.299 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0577 (1.0621) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.309 (0.306) data 0.000 (0.007) eta 0:32:23 loss 1.0846 (1.0768) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.248, TIME@all 0.305 -epoch: [223/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0871 (1.0658) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.308 (0.306) data 0.000 (0.006) eta 0:32:23 loss 1.1732 (1.0789) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.093, TIME@all 0.305 -epoch: [223/350][20/50] time 0.298 (0.305) data 0.000 (0.013) eta 0:32:24 loss 1.0736 (1.0615) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.305 (0.306) data 0.001 (0.007) eta 0:32:22 loss 1.0982 (1.0751) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 839.516, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.012) eta 0:32:15 loss 1.0589 (1.0715) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:32:08 loss 1.0985 (1.0884) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.133, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0697 (1.0706) acc 100.0000 (99.8438) lr 0.026000 -epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.1222 (1.0823) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.169, TIME@all 0.305 -epoch: [224/350][20/50] time 0.300 (0.305) data 0.000 (0.013) eta 0:32:10 loss 1.0873 (1.0775) acc 100.0000 (99.8438) lr 0.026000 -epoch: [224/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:32:06 loss 1.1253 (1.0812) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 839.859, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.014) eta 0:32:15 loss 1.0685 (1.0773) acc 100.0000 (99.6875) lr 0.026000 -epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.1036 (1.0842) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 839.270, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.014) eta 0:32:15 loss 1.0675 (1.0646) acc 100.0000 (99.8438) lr 0.026000 -epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.1082 (1.0787) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 839.331, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0787 (1.0655) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.0879 (1.0822) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 839.156, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:15 loss 1.0668 (1.0692) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:32:08 loss 1.1010 (1.0797) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.148, TIME@all 0.305 -epoch: [224/350][20/50] time 0.301 (0.306) data 0.000 (0.013) eta 0:32:14 loss 1.1143 (1.0768) acc 96.8750 (99.5312) lr 0.026000 -epoch: [224/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:32:07 loss 1.0911 (1.0808) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 839.511, TIME@all 0.305 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.0595 (1.0711) acc 100.0000 (99.8438) lr 0.026000 -epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.1035 (1.0860) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 836.801, TIME@all 0.306 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.1136 (1.0647) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.0814 (1.0719) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 836.828, TIME@all 0.306 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0696 (1.0622) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.304 (0.306) data 0.001 (0.007) eta 0:31:58 loss 1.0720 (1.0801) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 836.855, TIME@all 0.306 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.1251 (1.0682) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:31:57 loss 1.0897 (1.0791) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 836.970, TIME@all 0.306 -epoch: [225/350][20/50] time 0.300 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0630 (1.0633) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.303 (0.306) data 0.000 (0.007) eta 0:31:58 loss 1.0632 (1.0786) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 837.157, TIME@all 0.306 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.1726 (1.0669) acc 96.8750 (99.8438) lr 0.026000 -epoch: [225/350][40/50] time 0.305 (0.306) data 0.000 (0.007) eta 0:31:57 loss 1.1251 (1.0791) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 837.007, TIME@all 0.306 -epoch: [225/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:32:05 loss 1.0557 (1.0659) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:31:58 loss 1.0921 (1.0866) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 836.835, TIME@all 0.306 -epoch: [225/350][20/50] time 0.300 (0.307) data 0.000 (0.012) eta 0:32:05 loss 1.0599 (1.0651) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.304 (0.306) data 0.000 (0.006) eta 0:31:58 loss 1.0951 (1.0837) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 836.894, TIME@all 0.306 -epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:31:50 loss 1.1594 (1.0733) acc 96.8750 (99.8438) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.006) eta 0:31:40 loss 1.0714 (1.0754) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.529, TIME@all 0.305 -epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0868 (1.0671) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.320 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0514 (1.0774) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.606, TIME@all 0.305 -epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0966 (1.0651) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0761 (1.0763) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.549, TIME@all 0.305 -epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0957 (1.0690) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0916 (1.0783) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.541, TIME@all 0.305 -epoch: [226/350][20/50] time 0.304 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.0911 (1.0715) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:39 loss 1.1611 (1.0787) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.703, TIME@all 0.305 -epoch: [226/350][20/50] time 0.303 (0.307) data 0.000 (0.013) eta 0:31:49 loss 1.1367 (1.0691) acc 96.8750 (99.8438) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.306) data 0.000 (0.007) eta 0:31:39 loss 1.0959 (1.0807) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 838.939, TIME@all 0.305 -epoch: [226/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:31:50 loss 1.1645 (1.0660) acc 96.8750 (99.8438) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.000 (0.007) eta 0:31:40 loss 1.0686 (1.0764) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 838.542, TIME@all 0.305 -epoch: [226/350][20/50] time 0.304 (0.307) data 0.001 (0.014) eta 0:31:50 loss 1.1560 (1.0688) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.317 (0.306) data 0.001 (0.007) eta 0:31:39 loss 1.0925 (1.0800) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.744, TIME@all 0.305 -epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0731 (1.0640) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0734 (1.0678) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.632, TIME@all 0.305 -epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.012) eta 0:31:40 loss 1.0566 (1.0658) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:31:25 loss 1.0551 (1.0739) acc 100.0000 (99.4531) lr 0.002600 -FPS@all 838.625, TIME@all 0.305 -epoch: [227/350][20/50] time 0.334 (0.307) data 0.000 (0.012) eta 0:31:40 loss 1.0607 (1.0706) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.006) eta 0:31:25 loss 1.0866 (1.0724) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.706, TIME@all 0.305 -epoch: [227/350][20/50] time 0.335 (0.307) data 0.000 (0.014) eta 0:31:40 loss 1.0682 (1.0681) acc 100.0000 (99.6875) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:24 loss 1.0831 (1.0750) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 838.813, TIME@all 0.305 -epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0969 (1.0656) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0582 (1.0694) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.617, TIME@all 0.305 -epoch: [227/350][20/50] time 0.335 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.1015 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:24 loss 1.0708 (1.0674) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.772, TIME@all 0.305 -epoch: [227/350][20/50] time 0.334 (0.308) data 0.001 (0.013) eta 0:31:40 loss 1.1135 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.302 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0583 (1.0698) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.628, TIME@all 0.305 -epoch: [227/350][20/50] time 0.334 (0.308) data 0.000 (0.013) eta 0:31:40 loss 1.0772 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.301 (0.306) data 0.000 (0.007) eta 0:31:25 loss 1.0637 (1.0713) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.923, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:31:11 loss 1.0697 (1.0616) acc 100.0000 (99.8438) lr 0.002600 -epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0901 (1.0709) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.216, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 0:31:10 loss 1.0547 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0652 (1.0663) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.181, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:31:10 loss 1.0739 (1.0675) acc 100.0000 (99.6875) lr 0.002600 -epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:31:08 loss 1.1467 (1.0760) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 838.114, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.014) eta 0:31:11 loss 1.0570 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0733 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.145, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.013) eta 0:31:11 loss 1.0521 (1.0621) acc 100.0000 (99.8438) lr 0.002600 -epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0828 (1.0769) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 838.138, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.001 (0.014) eta 0:31:10 loss 1.0678 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.307 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0852 (1.0703) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.322, TIME@all 0.305 -epoch: [228/350][20/50] time 0.301 (0.305) data 0.000 (0.014) eta 0:31:10 loss 1.0841 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0606 (1.0772) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.287, TIME@all 0.305 -epoch: [228/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:31:10 loss 1.0642 (1.0607) acc 100.0000 (99.8438) lr 0.002600 -epoch: [228/350][40/50] time 0.308 (0.306) data 0.000 (0.007) eta 0:31:08 loss 1.0561 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.439, TIME@all 0.305 -epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:30:57 loss 1.0811 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0495 (1.0620) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 837.116, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.012) eta 0:30:57 loss 1.1091 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0478 (1.0718) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 837.091, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.011) eta 0:30:57 loss 1.1131 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0648 (1.0744) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 836.999, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:30:57 loss 1.1164 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.1143 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 837.053, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.306) data 0.000 (0.013) eta 0:30:57 loss 1.1093 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.0508 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 837.013, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:30:57 loss 1.1275 (1.0637) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.001 (0.007) eta 0:30:55 loss 1.0647 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 837.212, TIME@all 0.306 -epoch: [229/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:30:57 loss 1.0646 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:55 loss 1.0517 (1.0716) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 837.168, TIME@all 0.306 -epoch: [229/350][20/50] time 0.313 (0.306) data 0.000 (0.012) eta 0:30:57 loss 1.1966 (1.0689) acc 96.8750 (99.8438) lr 0.002600 -epoch: [229/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:55 loss 1.0545 (1.0711) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 837.357, TIME@all 0.306 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.011) eta 0:30:35 loss 1.1502 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0655 (1.0739) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.128, TIME@all 0.305 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1663 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0784 (1.0726) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 839.174, TIME@all 0.305 -epoch: [230/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1142 (1.0708) acc 100.0000 (99.8438) lr 0.002600 -epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0920 (1.0777) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.221, TIME@all 0.305 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:30:35 loss 1.1536 (1.0663) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.007) eta 0:30:36 loss 1.1164 (1.0745) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 839.156, TIME@all 0.305 -epoch: [230/350][20/50] time 0.301 (0.304) data 0.001 (0.012) eta 0:30:35 loss 1.1429 (1.0763) acc 100.0000 (99.6875) lr 0.002600 -epoch: [230/350][40/50] time 0.306 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0840 (1.0818) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 839.157, TIME@all 0.305 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:30:35 loss 1.1021 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.305 (0.305) data 0.001 (0.007) eta 0:30:35 loss 1.0822 (1.0755) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.349, TIME@all 0.305 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:30:35 loss 1.1059 (1.0693) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.0657 (1.0775) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 839.267, TIME@all 0.305 -epoch: [230/350][20/50] time 0.300 (0.304) data 0.001 (0.012) eta 0:30:35 loss 1.1046 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.305 (0.306) data 0.000 (0.006) eta 0:30:36 loss 1.1221 (1.0768) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 839.299, TIME@all 0.305 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.012) eta 0:30:37 loss 1.0546 (1.0658) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.006) eta 0:30:28 loss 1.0562 (1.0730) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 836.360, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0519 (1.0599) acc 100.0000 (99.8438) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0522 (1.0768) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 836.512, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0587 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.001 (0.007) eta 0:30:28 loss 1.0522 (1.0745) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 836.367, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0541 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0802 (1.0723) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 836.435, TIME@all 0.306 -epoch: [231/350][20/50] time 0.301 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0855 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.302 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0521 (1.0699) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 836.381, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:37 loss 1.0503 (1.0617) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0675 (1.0745) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 836.370, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:36 loss 1.0557 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.303 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.1006 (1.0725) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 836.564, TIME@all 0.306 -epoch: [231/350][20/50] time 0.302 (0.307) data 0.000 (0.013) eta 0:30:36 loss 1.0561 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.301 (0.307) data 0.000 (0.007) eta 0:30:27 loss 1.0471 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 836.738, TIME@all 0.306 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:23 loss 1.0676 (1.0596) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0716 (1.0644) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.414, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.012) eta 0:30:22 loss 1.0512 (1.0601) acc 100.0000 (99.8438) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.006) eta 0:30:11 loss 1.0624 (1.0676) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.377, TIME@all 0.305 -epoch: [232/350][20/50] time 0.304 (0.307) data 0.000 (0.012) eta 0:30:22 loss 1.0639 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.325 (0.307) data 0.000 (0.006) eta 0:30:11 loss 1.0913 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.491, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.014) eta 0:30:22 loss 1.0853 (1.0597) acc 100.0000 (99.8438) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.306) data 0.000 (0.007) eta 0:30:11 loss 1.0514 (1.0619) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.614, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0590 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0798 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.385, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0924 (1.0612) acc 96.8750 (99.8438) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0534 (1.0631) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.430, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:22 loss 1.0801 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.322 (0.306) data 0.000 (0.007) eta 0:30:11 loss 1.0584 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 838.556, TIME@all 0.305 -epoch: [232/350][20/50] time 0.305 (0.307) data 0.000 (0.013) eta 0:30:23 loss 1.0841 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.324 (0.307) data 0.000 (0.007) eta 0:30:11 loss 1.0857 (1.0639) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 838.652, TIME@all 0.305 -epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:29:49 loss 1.0794 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0489 (1.0722) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.839, TIME@all 0.304 -epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:49 loss 1.0637 (1.0603) acc 100.0000 (99.8438) lr 0.002600 -epoch: [233/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0529 (1.0685) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.866, TIME@all 0.304 -epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:29:48 loss 1.0836 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0628 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.932, TIME@all 0.304 -epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:49 loss 1.0668 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0547 (1.0724) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.891, TIME@all 0.304 -epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:49 loss 1.0542 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:29:43 loss 1.0552 (1.0756) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.846, TIME@all 0.304 -epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:48 loss 1.0598 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:29:42 loss 1.0473 (1.0720) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.046, TIME@all 0.304 -epoch: [233/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:29:48 loss 1.0614 (1.0655) acc 100.0000 (99.6875) lr 0.002600 -epoch: [233/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:29:42 loss 1.0634 (1.0749) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.006, TIME@all 0.304 -epoch: [233/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:29:48 loss 1.0649 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -epoch: [233/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:29:43 loss 1.0696 (1.0702) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.221, TIME@all 0.304 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0660 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0754 (1.0692) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.455, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0589 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0580 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.427, TIME@all 0.303 -epoch: [234/350][20/50] time 0.301 (0.302) data 0.000 (0.012) eta 0:29:22 loss 1.0538 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:29:22 loss 1.0557 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.466, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0602 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0539 (1.0629) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.419, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0566 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0493 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.559, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0626 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0619 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.426, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.014) eta 0:29:22 loss 1.0579 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0507 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.631, TIME@all 0.303 -epoch: [234/350][20/50] time 0.300 (0.302) data 0.000 (0.013) eta 0:29:22 loss 1.0531 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:29:22 loss 1.0510 (1.0644) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.828, TIME@all 0.303 -epoch: [235/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:29:16 loss 1.0606 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.0514 (1.0611) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.555, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:29:16 loss 1.0779 (1.0595) acc 100.0000 (99.8438) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.0585 (1.0693) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.618, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.1027 (1.0549) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0760 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.560, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.1783 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.1281 (1.0711) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.586, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.1109 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:08 loss 1.0725 (1.0674) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.775, TIME@all 0.303 -epoch: [235/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.0944 (1.0545) acc 96.8750 (99.8438) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:29:09 loss 1.1023 (1.0638) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.917, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:15 loss 1.0698 (1.0519) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0710 (1.0628) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.719, TIME@all 0.303 -epoch: [235/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:29:16 loss 1.0480 (1.0525) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:29:09 loss 1.0593 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.562, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:28:59 loss 1.0690 (1.0587) acc 100.0000 (99.8438) lr 0.002600 -epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:53 loss 1.0535 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.330, TIME@all 0.303 -epoch: [236/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:29:00 loss 1.0778 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0736 (1.0719) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.312, TIME@all 0.303 -epoch: [236/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:28:59 loss 1.0646 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:53 loss 1.0536 (1.0737) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.423, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0958 (1.0631) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0504 (1.0703) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.543, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0964 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0512 (1.0702) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.327, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0558 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:28:53 loss 1.0780 (1.0683) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.337, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:59 loss 1.0823 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0621 (1.0710) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.494, TIME@all 0.303 -epoch: [236/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:28:59 loss 1.0801 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:28:53 loss 1.0713 (1.0704) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.640, TIME@all 0.303 -epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.1111 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0725 (1.0739) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.775, TIME@all 0.304 -epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:28:53 loss 1.0745 (1.0682) acc 100.0000 (99.8438) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0619 (1.0757) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.848, TIME@all 0.304 -epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.0845 (1.0572) acc 100.0000 (99.8438) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1064 (1.0693) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.690, TIME@all 0.304 -epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 0:28:54 loss 1.0766 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1097 (1.0724) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.716, TIME@all 0.304 -epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:28:53 loss 1.1845 (1.0620) acc 96.8750 (99.8438) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1033 (1.0731) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 841.939, TIME@all 0.304 -epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:28:54 loss 1.0957 (1.0653) acc 100.0000 (99.8438) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.1526 (1.0720) acc 96.8750 (99.6094) lr 0.002600 -FPS@all 841.715, TIME@all 0.304 -epoch: [237/350][20/50] time 0.308 (0.305) data 0.000 (0.014) eta 0:28:54 loss 1.1196 (1.0688) acc 100.0000 (99.6875) lr 0.002600 -epoch: [237/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0943 (1.0708) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.108, TIME@all 0.304 -epoch: [237/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:28:53 loss 1.0854 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:28:44 loss 1.0548 (1.0712) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.851, TIME@all 0.304 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0523 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0640 (1.0709) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.160, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0537 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0483 (1.0680) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.091, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:32 loss 1.0691 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0571 (1.0746) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.200, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:28:32 loss 1.0745 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:25 loss 1.0522 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.365, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.001 (0.012) eta 0:28:32 loss 1.0656 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0690 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.545, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:32 loss 1.0582 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0600 (1.0724) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.167, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:28:33 loss 1.0565 (1.0578) acc 100.0000 (99.8438) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:28:25 loss 1.0504 (1.0642) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.136, TIME@all 0.303 -epoch: [238/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:28:32 loss 1.0756 (1.0616) acc 100.0000 (99.8438) lr 0.002600 -epoch: [238/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:28:25 loss 1.0660 (1.0707) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.322, TIME@all 0.303 -epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.011) eta 0:28:16 loss 1.0573 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0578 (1.0671) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.790, TIME@all 0.304 -epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0596 (1.0561) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0551 (1.0682) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.832, TIME@all 0.304 -epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0569 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0506 (1.0719) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.841, TIME@all 0.304 -epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:15 loss 1.0488 (1.0638) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:28:10 loss 1.0572 (1.0739) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.987, TIME@all 0.304 -epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:28:16 loss 1.0617 (1.0620) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0607 (1.0699) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.817, TIME@all 0.304 -epoch: [239/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:28:16 loss 1.0530 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0509 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.808, TIME@all 0.304 -epoch: [239/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:28:15 loss 1.0568 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:28:10 loss 1.0491 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.976, TIME@all 0.304 -epoch: [239/350][20/50] time 0.299 (0.304) data 0.000 (0.012) eta 0:28:15 loss 1.0490 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:28:10 loss 1.0495 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.258, TIME@all 0.304 -epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.012) eta 0:28:07 loss 1.0766 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:27:59 loss 1.1132 (1.0669) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 842.019, TIME@all 0.304 -epoch: [240/350][20/50] time 0.321 (0.305) data 0.000 (0.012) eta 0:28:08 loss 1.0560 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.006) eta 0:27:58 loss 1.0743 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.140, TIME@all 0.304 -epoch: [240/350][20/50] time 0.325 (0.306) data 0.000 (0.013) eta 0:28:09 loss 1.0664 (1.0658) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.307 (0.305) data 0.000 (0.007) eta 0:27:59 loss 1.1516 (1.0700) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 842.073, TIME@all 0.304 -epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0803 (1.0684) acc 100.0000 (99.6875) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:59 loss 1.0778 (1.0678) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.030, TIME@all 0.304 -epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.014) eta 0:28:06 loss 1.1427 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0695 (1.0696) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.248, TIME@all 0.304 -epoch: [240/350][20/50] time 0.319 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0599 (1.0689) acc 100.0000 (99.6875) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0865 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.384, TIME@all 0.304 -epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:06 loss 1.1005 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.000 (0.007) eta 0:27:58 loss 1.0954 (1.0709) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.193, TIME@all 0.304 -epoch: [240/350][20/50] time 0.318 (0.305) data 0.000 (0.013) eta 0:28:07 loss 1.0707 (1.0630) acc 100.0000 (99.8438) lr 0.002600 -epoch: [240/350][40/50] time 0.306 (0.305) data 0.001 (0.007) eta 0:27:59 loss 1.2014 (1.0689) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 842.041, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0675 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.2126 (1.0694) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 842.412, TIME@all 0.304 -epoch: [241/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0580 (1.0532) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0818 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.458, TIME@all 0.304 -epoch: [241/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.0745 (1.0556) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0595 (1.0664) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.495, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0721 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0691 (1.0713) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.415, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0579 (1.0546) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0615 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.641, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:47 loss 1.1259 (1.0616) acc 96.8750 (99.8438) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:27:41 loss 1.0517 (1.0719) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.423, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0646 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0900 (1.0655) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.773, TIME@all 0.304 -epoch: [241/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:27:47 loss 1.0586 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:27:41 loss 1.0508 (1.0671) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.549, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:27:32 loss 1.0602 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:27 loss 1.1332 (1.0653) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.036, TIME@all 0.304 -epoch: [242/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:31 loss 1.0726 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.1061 (1.0677) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.087, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:27:31 loss 1.0500 (1.0604) acc 100.0000 (99.8438) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.1129 (1.0633) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.120, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:31 loss 1.0644 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.006) eta 0:27:26 loss 1.0629 (1.0639) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.043, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:32 loss 1.0574 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:27 loss 1.0683 (1.0688) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.051, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:27:31 loss 1.0881 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.0963 (1.0654) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.238, TIME@all 0.304 -epoch: [242/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:27:31 loss 1.0757 (1.0596) acc 100.0000 (99.6875) lr 0.002600 -epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.1029 (1.0706) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 842.201, TIME@all 0.304 -epoch: [242/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:27:31 loss 1.0554 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.316 (0.304) data 0.000 (0.007) eta 0:27:26 loss 1.1041 (1.0622) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.413, TIME@all 0.304 -epoch: [243/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.0680 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0529 (1.0669) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.522, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.1270 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0539 (1.0681) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.613, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0735 (1.0637) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0575 (1.0701) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.548, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0662 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0516 (1.0696) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.555, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:13 loss 1.0554 (1.0570) acc 100.0000 (99.8438) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0708 (1.0640) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.743, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:27:14 loss 1.0592 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:27:10 loss 1.0629 (1.0698) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.562, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0572 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0570 (1.0679) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.692, TIME@all 0.304 -epoch: [243/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:27:14 loss 1.0582 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:27:10 loss 1.0658 (1.0700) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.909, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0542 (1.0521) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0503 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.584, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.011) eta 0:27:03 loss 1.0894 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0839 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.664, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0748 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0835 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.576, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0504 (1.0514) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0564 (1.0601) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.591, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:27:03 loss 1.0925 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:56 loss 1.0597 (1.0680) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.622, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:02 loss 1.0482 (1.0518) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0797 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.991, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:27:02 loss 1.0543 (1.0533) acc 100.0000 (99.8438) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:56 loss 1.0697 (1.0628) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.735, TIME@all 0.304 -epoch: [244/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:27:02 loss 1.0674 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:26:56 loss 1.0711 (1.0620) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.796, TIME@all 0.304 -epoch: [245/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0496 (1.0723) acc 100.0000 (99.6875) lr 0.002600 -epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0612 (1.0762) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.626, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0531 (1.0631) acc 100.0000 (99.8438) lr 0.002600 -epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0560 (1.0699) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.665, TIME@all 0.304 -epoch: [245/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0500 (1.0658) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0650 (1.0673) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.677, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0514 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0590 (1.0685) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.825, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0504 (1.0598) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:26:41 loss 1.0801 (1.0711) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.632, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0571 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0615 (1.0680) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.624, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.014) eta 0:26:48 loss 1.0722 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0565 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.791, TIME@all 0.304 -epoch: [245/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:26:48 loss 1.0562 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:26:41 loss 1.0545 (1.0703) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.944, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.012) eta 0:26:30 loss 1.0540 (1.0609) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:25 loss 1.0658 (1.0690) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.181, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:26:30 loss 1.0564 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:25 loss 1.0667 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.208, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:26:30 loss 1.0477 (1.0673) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:25 loss 1.0588 (1.0710) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.209, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:26:30 loss 1.0602 (1.0600) acc 100.0000 (99.8438) lr 0.002600 -epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:24 loss 1.0520 (1.0729) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.408, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.013) eta 0:26:30 loss 1.0604 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:25 loss 1.0909 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.214, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.001 (0.012) eta 0:26:30 loss 1.0583 (1.0615) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:26:24 loss 1.0700 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.251, TIME@all 0.304 -epoch: [246/350][20/50] time 0.310 (0.304) data 0.000 (0.013) eta 0:26:30 loss 1.0921 (1.0717) acc 100.0000 (99.5312) lr 0.002600 -epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:26:24 loss 1.0718 (1.0711) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.367, TIME@all 0.304 -epoch: [246/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:26:30 loss 1.1025 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:26:24 loss 1.0799 (1.0704) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.613, TIME@all 0.303 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:26:19 loss 1.0700 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.1010 (1.0678) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 840.781, TIME@all 0.304 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.011) eta 0:26:19 loss 1.1485 (1.0696) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0757 (1.0700) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 840.811, TIME@all 0.304 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0522 (1.0563) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0581 (1.0624) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 840.953, TIME@all 0.304 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0714 (1.0549) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0693 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 840.778, TIME@all 0.304 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:26:19 loss 1.0884 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.007) eta 0:26:13 loss 1.0634 (1.0723) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.003, TIME@all 0.304 -epoch: [247/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.0870 (1.0631) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.1297 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 840.802, TIME@all 0.304 -epoch: [247/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.1296 (1.0650) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0552 (1.0674) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.102, TIME@all 0.304 -epoch: [247/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:26:19 loss 1.2360 (1.0637) acc 96.8750 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:26:13 loss 1.0858 (1.0703) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 840.828, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:25:58 loss 1.0452 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:25:53 loss 1.0537 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.056, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0517 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0831 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.118, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:57 loss 1.0450 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:52 loss 1.0692 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.089, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:57 loss 1.0481 (1.0522) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:25:52 loss 1.0498 (1.0682) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.263, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:58 loss 1.0559 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0458 (1.0615) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.059, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:25:57 loss 1.0702 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:52 loss 1.0579 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.218, TIME@all 0.304 -epoch: [248/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0488 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0542 (1.0678) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.076, TIME@all 0.304 -epoch: [248/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:25:58 loss 1.0499 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:53 loss 1.0530 (1.0665) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.372, TIME@all 0.304 -epoch: [249/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.0646 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:39 loss 1.0790 (1.0654) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.725, TIME@all 0.303 -epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:25:46 loss 1.0560 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:38 loss 1.0510 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.775, TIME@all 0.303 -epoch: [249/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:25:46 loss 1.0851 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:39 loss 1.0547 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.691, TIME@all 0.303 -epoch: [249/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:25:46 loss 1.0856 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:39 loss 1.0603 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.723, TIME@all 0.303 -epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.1087 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:39 loss 1.0586 (1.0637) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.734, TIME@all 0.303 -epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.0564 (1.0609) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:38 loss 1.0740 (1.0718) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.918, TIME@all 0.303 -epoch: [249/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.2058 (1.0704) acc 96.8750 (99.6875) lr 0.002600 -epoch: [249/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:25:39 loss 1.0546 (1.0744) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.987, TIME@all 0.303 -epoch: [249/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:46 loss 1.1066 (1.0581) acc 96.8750 (99.8438) lr 0.002600 -epoch: [249/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:25:38 loss 1.1457 (1.0695) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.846, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.011) eta 0:25:27 loss 1.0620 (1.0636) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0700 (1.0718) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.563, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0702 (1.0671) acc 100.0000 (99.6875) lr 0.002600 -epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0596 (1.0749) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 843.571, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0827 (1.0606) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:25:24 loss 1.1223 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.581, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0488 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.1120 (1.0728) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.620, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:25:27 loss 1.0987 (1.0655) acc 100.0000 (99.6875) lr 0.002600 -epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:25:23 loss 1.0833 (1.0719) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.737, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0694 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:24 loss 1.0663 (1.0693) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.572, TIME@all 0.303 -epoch: [250/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:25:27 loss 1.0676 (1.0617) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:25:23 loss 1.0622 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.696, TIME@all 0.303 -epoch: [250/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:25:26 loss 1.0697 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:25:23 loss 1.0694 (1.0695) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.989, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0743 (1.0581) acc 100.0000 (99.8438) lr 0.002600 -epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.0940 (1.0701) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.013, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:25:14 loss 1.1017 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:25:08 loss 1.0687 (1.0705) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.077, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1298 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.0604 (1.0731) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.976, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1059 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1003 (1.0709) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 844.002, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.1578 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1072 (1.0663) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 844.017, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0864 (1.0595) acc 100.0000 (99.8438) lr 0.002600 -epoch: [251/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:25:07 loss 1.0921 (1.0702) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.197, TIME@all 0.303 -epoch: [251/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0825 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:25:08 loss 1.0544 (1.0681) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.338, TIME@all 0.303 -epoch: [251/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:25:14 loss 1.0809 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:25:08 loss 1.1144 (1.0704) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.131, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0785 (1.0610) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0625 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.679, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0918 (1.0633) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.300 (0.304) data 0.001 (0.006) eta 0:24:52 loss 1.0472 (1.0632) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.719, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:24:59 loss 1.0556 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0503 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.759, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0664 (1.0590) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0512 (1.0637) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.883, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0954 (1.0563) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:24:52 loss 1.0522 (1.0612) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.676, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.1267 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0423 (1.0669) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.830, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0785 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:24:52 loss 1.0486 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.034, TIME@all 0.303 -epoch: [252/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:24:59 loss 1.0514 (1.0578) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:24:52 loss 1.0452 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.655, TIME@all 0.303 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:48 loss 1.1000 (1.0593) acc 100.0000 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0431 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.196, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:48 loss 1.1287 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0627 (1.0648) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.250, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.1076 (1.0586) acc 100.0000 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0566 (1.0646) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.360, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.0866 (1.0592) acc 100.0000 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0855 (1.0630) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.292, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:49 loss 1.1651 (1.0586) acc 96.8750 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:38 loss 1.0469 (1.0616) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.221, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:49 loss 1.0895 (1.0546) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0749 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.216, TIME@all 0.304 -epoch: [253/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:24:48 loss 1.0983 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:38 loss 1.0994 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.417, TIME@all 0.304 -epoch: [253/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:24:49 loss 1.0814 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:24:38 loss 1.0464 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.525, TIME@all 0.303 -epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0706 (1.0631) acc 100.0000 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1212 (1.0723) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.021, TIME@all 0.304 -epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0991 (1.0598) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1128 (1.0712) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.946, TIME@all 0.304 -epoch: [254/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:24:27 loss 1.1010 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.1191 (1.0697) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.914, TIME@all 0.304 -epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:24:27 loss 1.0837 (1.0560) acc 100.0000 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:24:21 loss 1.1003 (1.0622) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.131, TIME@all 0.304 -epoch: [254/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:24:28 loss 1.0613 (1.0545) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0562 (1.0705) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.150, TIME@all 0.304 -epoch: [254/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:24:27 loss 1.0720 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0990 (1.0702) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.946, TIME@all 0.304 -epoch: [254/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0709 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:21 loss 1.1047 (1.0627) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 843.073, TIME@all 0.304 -epoch: [254/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:24:27 loss 1.0522 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:24:22 loss 1.0478 (1.0635) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.927, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:24:14 loss 1.0631 (1.0603) acc 100.0000 (99.8438) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:24:08 loss 1.0787 (1.0631) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.014, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.001 (0.013) eta 0:24:14 loss 1.0643 (1.0617) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0924 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.076, TIME@all 0.304 -epoch: [255/350][20/50] time 0.301 (0.304) data 0.001 (0.013) eta 0:24:14 loss 1.1146 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.1478 (1.0669) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.093, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:24:14 loss 1.0637 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:07 loss 1.0509 (1.0631) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.244, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0622 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0903 (1.0642) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.033, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0902 (1.0611) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0646 (1.0658) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.030, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:24:14 loss 1.0503 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:07 loss 1.0771 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.194, TIME@all 0.304 -epoch: [255/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:24:14 loss 1.0756 (1.0621) acc 100.0000 (99.8438) lr 0.002600 -epoch: [255/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:24:08 loss 1.0799 (1.0627) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.407, TIME@all 0.304 -epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:01 loss 1.1149 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0679 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.651, TIME@all 0.304 -epoch: [256/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0460 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0548 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.668, TIME@all 0.304 -epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0694 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0625 (1.0653) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.731, TIME@all 0.304 -epoch: [256/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:24:00 loss 1.0964 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:23:53 loss 1.0592 (1.0645) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.672, TIME@all 0.304 -epoch: [256/350][20/50] time 0.306 (0.305) data 0.001 (0.012) eta 0:24:00 loss 1.0632 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0615 (1.0616) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.670, TIME@all 0.304 -epoch: [256/350][20/50] time 0.306 (0.305) data 0.001 (0.013) eta 0:24:00 loss 1.0604 (1.0501) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:53 loss 1.0453 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.854, TIME@all 0.304 -epoch: [256/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:24:00 loss 1.0820 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0566 (1.0652) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.807, TIME@all 0.304 -epoch: [256/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:24:01 loss 1.0806 (1.0595) acc 100.0000 (99.8438) lr 0.002600 -epoch: [256/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:53 loss 1.0484 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.822, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:23:41 loss 1.0540 (1.0594) acc 100.0000 (99.8438) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:37 loss 1.0836 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.964, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:23:41 loss 1.1200 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:23:37 loss 1.0904 (1.0650) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.000, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0997 (1.0625) acc 100.0000 (99.6875) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0490 (1.0695) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.009, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.0953 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0572 (1.0659) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.196, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.1136 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0802 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.998, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:41 loss 1.0566 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0636 (1.0672) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.968, TIME@all 0.304 -epoch: [257/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0957 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.0677 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.146, TIME@all 0.304 -epoch: [257/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:41 loss 1.0965 (1.0593) acc 96.8750 (99.6875) lr 0.002600 -epoch: [257/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:37 loss 1.1397 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.338, TIME@all 0.304 -epoch: [258/350][20/50] time 0.302 (0.305) data 0.000 (0.011) eta 0:23:33 loss 1.1860 (1.0625) acc 96.8750 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0890 (1.0713) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.788, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0677 (1.0607) acc 100.0000 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.302 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0636 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.849, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.1086 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.1014 (1.0680) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.835, TIME@all 0.304 -epoch: [258/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:23:32 loss 1.1303 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0634 (1.0679) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.871, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:23:32 loss 1.1075 (1.0609) acc 100.0000 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.001 (0.007) eta 0:23:25 loss 1.1182 (1.0702) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 842.016, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0732 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0613 (1.0684) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.160, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:23:33 loss 1.0736 (1.0567) acc 100.0000 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.001 (0.006) eta 0:23:25 loss 1.0652 (1.0624) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.842, TIME@all 0.304 -epoch: [258/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:23:32 loss 1.0589 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.303 (0.305) data 0.000 (0.006) eta 0:23:25 loss 1.0770 (1.0670) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.971, TIME@all 0.304 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1125 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0465 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.617, TIME@all 0.303 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1359 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0741 (1.0687) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.587, TIME@all 0.303 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.0805 (1.0569) acc 100.0000 (99.8438) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:23:05 loss 1.0498 (1.0639) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.583, TIME@all 0.303 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:23:10 loss 1.1501 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:23:05 loss 1.0560 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.783, TIME@all 0.303 -epoch: [259/350][20/50] time 0.300 (0.304) data 0.001 (0.013) eta 0:23:10 loss 1.0602 (1.0571) acc 100.0000 (99.8438) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0552 (1.0701) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.614, TIME@all 0.303 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.1367 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0523 (1.0692) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.608, TIME@all 0.303 -epoch: [259/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:23:10 loss 1.1268 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0465 (1.0712) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.752, TIME@all 0.303 -epoch: [259/350][20/50] time 0.300 (0.304) data 0.000 (0.013) eta 0:23:11 loss 1.0776 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:23:05 loss 1.0586 (1.0648) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.775, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.1067 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0513 (1.0706) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.502, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:23:00 loss 1.0882 (1.0608) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:22:52 loss 1.0473 (1.0666) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.508, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.0799 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0490 (1.0732) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.588, TIME@all 0.303 -epoch: [260/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.1094 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0485 (1.0674) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.515, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:22:59 loss 1.0516 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:52 loss 1.0492 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.543, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:22:59 loss 1.0580 (1.0611) acc 100.0000 (99.6875) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:22:51 loss 1.1054 (1.0713) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.724, TIME@all 0.303 -epoch: [260/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:22:59 loss 1.0908 (1.0579) acc 100.0000 (99.8438) lr 0.002600 -epoch: [260/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:22:51 loss 1.0482 (1.0720) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.862, TIME@all 0.303 -epoch: [260/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:22:59 loss 1.0716 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:22:51 loss 1.0572 (1.0644) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.658, TIME@all 0.303 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0617 (1.0630) acc 100.0000 (99.8438) lr 0.002600 -epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0810 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 846.116, TIME@all 0.303 -epoch: [261/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0763 (1.0580) acc 100.0000 (99.8438) lr 0.002600 -epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0731 (1.0691) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 846.211, TIME@all 0.303 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0699 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:22:30 loss 1.0580 (1.0661) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 846.267, TIME@all 0.303 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0799 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0705 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 846.093, TIME@all 0.303 -epoch: [261/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0483 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0816 (1.0703) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 846.120, TIME@all 0.303 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0870 (1.0556) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0765 (1.0672) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 846.111, TIME@all 0.303 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:22:38 loss 1.0648 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:22:30 loss 1.0686 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 846.353, TIME@all 0.302 -epoch: [261/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:22:38 loss 1.0732 (1.0530) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:22:30 loss 1.0682 (1.0647) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 846.317, TIME@all 0.302 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0481 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0591 (1.0613) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.959, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0572 (1.0596) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0633 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.999, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0878 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0737 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.008, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0450 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.1449 (1.0646) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.971, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0466 (1.0564) acc 100.0000 (99.8438) lr 0.002600 -epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0705 (1.0682) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.977, TIME@all 0.304 -epoch: [262/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0566 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0618 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.358, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:22:27 loss 1.0929 (1.0701) acc 100.0000 (99.6875) lr 0.002600 -epoch: [262/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:22:22 loss 1.0756 (1.0741) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.112, TIME@all 0.304 -epoch: [262/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:22:27 loss 1.0685 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:22:22 loss 1.0825 (1.0624) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.168, TIME@all 0.304 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:22:10 loss 1.0522 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:22:04 loss 1.0624 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.504, TIME@all 0.303 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0819 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.0639 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.543, TIME@all 0.303 -epoch: [263/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:22:10 loss 1.0488 (1.0532) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:05 loss 1.0567 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.481, TIME@all 0.304 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0490 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:05 loss 1.0756 (1.0697) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.473, TIME@all 0.304 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:22:09 loss 1.0515 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.1038 (1.0615) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.662, TIME@all 0.303 -epoch: [263/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0893 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.1016 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.541, TIME@all 0.303 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:22:10 loss 1.0481 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:22:05 loss 1.0758 (1.0678) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.721, TIME@all 0.303 -epoch: [263/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:22:09 loss 1.0754 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:22:04 loss 1.0852 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.624, TIME@all 0.303 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.011) eta 0:21:59 loss 1.0460 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0653 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.888, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0646 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0493 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.942, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0598 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0995 (1.0692) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.943, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0543 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0466 (1.0678) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.853, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0469 (1.0518) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0694 (1.0681) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.870, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.013) eta 0:21:59 loss 1.0536 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:50 loss 1.0488 (1.0653) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.083, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0494 (1.0521) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0571 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.043, TIME@all 0.304 -epoch: [264/350][20/50] time 0.301 (0.305) data 0.000 (0.012) eta 0:21:59 loss 1.0810 (1.0568) acc 100.0000 (99.8438) lr 0.002600 -epoch: [264/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:50 loss 1.0614 (1.0684) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.195, TIME@all 0.304 -epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:21:42 loss 1.0571 (1.0620) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:21:33 loss 1.0804 (1.0670) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.212, TIME@all 0.303 -epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0531 (1.0602) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0521 (1.0662) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.143, TIME@all 0.303 -epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0606 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0516 (1.0631) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.194, TIME@all 0.303 -epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0631 (1.0632) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0485 (1.0713) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.195, TIME@all 0.303 -epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0911 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0551 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.311, TIME@all 0.303 -epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0522 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0584 (1.0631) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.125, TIME@all 0.303 -epoch: [265/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:21:42 loss 1.0612 (1.0532) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:33 loss 1.0492 (1.0604) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.359, TIME@all 0.303 -epoch: [265/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:21:42 loss 1.0611 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:21:33 loss 1.0464 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.510, TIME@all 0.303 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:21:28 loss 1.0769 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0854 (1.0723) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.858, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1275 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.1036 (1.0695) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 841.797, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:21:28 loss 1.0653 (1.0601) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.1305 (1.0685) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.758, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1048 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.0991 (1.0664) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 841.771, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:21:28 loss 1.0784 (1.0556) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:22 loss 1.0724 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.968, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1257 (1.0702) acc 100.0000 (99.5312) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.007) eta 0:21:23 loss 1.1068 (1.0735) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 841.907, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:21:28 loss 1.1555 (1.0650) acc 100.0000 (99.8438) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0872 (1.0684) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 841.996, TIME@all 0.304 -epoch: [266/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:21:28 loss 1.0794 (1.0663) acc 100.0000 (99.6875) lr 0.002600 -epoch: [266/350][40/50] time 0.305 (0.305) data 0.000 (0.006) eta 0:21:23 loss 1.0614 (1.0695) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 841.788, TIME@all 0.304 -epoch: [267/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.1048 (1.0603) acc 96.8750 (99.6875) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0561 (1.0660) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.263, TIME@all 0.304 -epoch: [267/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0524 (1.0609) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:05 loss 1.0756 (1.0660) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.278, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:11 loss 1.0560 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:05 loss 1.0643 (1.0655) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.341, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0687 (1.0603) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0707 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.239, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0861 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:21:06 loss 1.0621 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.252, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:21:11 loss 1.0689 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:05 loss 1.0690 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.454, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:21:11 loss 1.0910 (1.0637) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:21:05 loss 1.0678 (1.0656) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.425, TIME@all 0.304 -epoch: [267/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:21:12 loss 1.0712 (1.0611) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:21:05 loss 1.1059 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.560, TIME@all 0.304 -epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0822 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0737 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 841.976, TIME@all 0.304 -epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0469 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0856 (1.0648) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.989, TIME@all 0.304 -epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0841 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0677 (1.0678) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.019, TIME@all 0.304 -epoch: [268/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.1118 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0758 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.011, TIME@all 0.304 -epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0715 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0567 (1.0631) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 841.991, TIME@all 0.304 -epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:20:59 loss 1.0791 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.300 (0.305) data 0.000 (0.006) eta 0:20:51 loss 1.0676 (1.0624) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.379, TIME@all 0.304 -epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0520 (1.0640) acc 100.0000 (99.8438) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0884 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.185, TIME@all 0.304 -epoch: [268/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:20:59 loss 1.0549 (1.0596) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.301 (0.305) data 0.000 (0.007) eta 0:20:51 loss 1.0965 (1.0653) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.133, TIME@all 0.304 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:20:40 loss 1.0763 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:20:32 loss 1.1087 (1.0731) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.690, TIME@all 0.303 -epoch: [269/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0659 (1.0578) acc 100.0000 (99.6875) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1135 (1.0717) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.713, TIME@all 0.303 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0629 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1002 (1.0717) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.696, TIME@all 0.303 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.0914 (1.0632) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0623 (1.0706) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.758, TIME@all 0.303 -epoch: [269/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:20:40 loss 1.1278 (1.0603) acc 96.8750 (99.6875) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0662 (1.0697) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.679, TIME@all 0.303 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:20:40 loss 1.0484 (1.0575) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1032 (1.0684) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.886, TIME@all 0.303 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.001 (0.013) eta 0:20:40 loss 1.0689 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.0780 (1.0675) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.000, TIME@all 0.303 -epoch: [269/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:20:40 loss 1.0495 (1.0580) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:20:32 loss 1.1145 (1.0671) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.833, TIME@all 0.303 -epoch: [270/350][20/50] time 0.297 (0.304) data 0.000 (0.011) eta 0:20:23 loss 1.0846 (1.0626) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0518 (1.0631) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.460, TIME@all 0.304 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0571 (1.0564) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0501 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.539, TIME@all 0.303 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:22 loss 1.0551 (1.0572) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0668 (1.0646) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.490, TIME@all 0.304 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:23 loss 1.0459 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:20:19 loss 1.0570 (1.0679) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.495, TIME@all 0.303 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:22 loss 1.0465 (1.0552) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:18 loss 1.0542 (1.0616) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.633, TIME@all 0.303 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0484 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0650 (1.0677) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.492, TIME@all 0.304 -epoch: [270/350][20/50] time 0.297 (0.303) data 0.000 (0.013) eta 0:20:22 loss 1.0578 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:20:18 loss 1.0620 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.686, TIME@all 0.303 -epoch: [270/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:20:23 loss 1.0675 (1.0501) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:20:19 loss 1.0962 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.827, TIME@all 0.303 -epoch: [271/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1264 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0698 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.278, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1049 (1.0637) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0662 (1.0753) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.311, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.0762 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.307 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.1155 (1.0746) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.330, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:14 loss 1.1351 (1.0642) acc 96.8750 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.1669 (1.0705) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 842.264, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:20:13 loss 1.1678 (1.0715) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:20:06 loss 1.0944 (1.0753) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.494, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.1332 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0688 (1.0741) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.281, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.013) eta 0:20:13 loss 1.1086 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.007) eta 0:20:06 loss 1.0906 (1.0757) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.429, TIME@all 0.304 -epoch: [271/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:20:13 loss 1.0917 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.308 (0.305) data 0.000 (0.006) eta 0:20:06 loss 1.0969 (1.0764) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.658, TIME@all 0.304 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0614 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0596 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.817, TIME@all 0.303 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0492 (1.0515) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0809 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.839, TIME@all 0.303 -epoch: [272/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:56 loss 1.0477 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:48 loss 1.0584 (1.0678) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.892, TIME@all 0.303 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.014) eta 0:19:56 loss 1.0579 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:19:48 loss 1.1278 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.006, TIME@all 0.303 -epoch: [272/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:19:56 loss 1.0674 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0495 (1.0640) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.019, TIME@all 0.303 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0429 (1.0511) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0637 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.822, TIME@all 0.303 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0633 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0665 (1.0659) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.972, TIME@all 0.303 -epoch: [272/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:19:56 loss 1.0593 (1.0493) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:48 loss 1.0537 (1.0622) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.803, TIME@all 0.303 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0812 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0631 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.135, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:40 loss 1.0699 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:19:34 loss 1.0633 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.052, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0731 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:19:34 loss 1.0563 (1.0632) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.160, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.0629 (1.0549) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0552 (1.0641) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.079, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:19:40 loss 1.0877 (1.0626) acc 100.0000 (99.6875) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:33 loss 1.0776 (1.0635) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.265, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:40 loss 1.1475 (1.0679) acc 96.8750 (99.5312) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0681 (1.0674) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.084, TIME@all 0.304 -epoch: [273/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:19:40 loss 1.0746 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:19:33 loss 1.0526 (1.0698) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.252, TIME@all 0.304 -epoch: [273/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:41 loss 1.0751 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:34 loss 1.0627 (1.0588) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.260, TIME@all 0.304 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0620 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:16 loss 1.0541 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.006, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.1743 (1.0624) acc 96.8750 (99.8438) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0619 (1.0692) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.884, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.011) eta 0:19:25 loss 1.0525 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0576 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.889, TIME@all 0.303 -epoch: [274/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0501 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:16 loss 1.0554 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.266, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0649 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0730 (1.0624) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.895, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:25 loss 1.0498 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:19:16 loss 1.0852 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.112, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:19:25 loss 1.0641 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:19:17 loss 1.0505 (1.0612) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.906, TIME@all 0.303 -epoch: [274/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:19:25 loss 1.0622 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:19:16 loss 1.0938 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.053, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0961 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:19:01 loss 1.0700 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.735, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.013) eta 0:19:05 loss 1.0860 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:02 loss 1.0572 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.680, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0836 (1.0520) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:19:02 loss 1.1468 (1.0645) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.658, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.011) eta 0:19:05 loss 1.0763 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:19:02 loss 1.0665 (1.0683) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.662, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.014) eta 0:19:05 loss 1.0668 (1.0599) acc 100.0000 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:01 loss 1.0502 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.843, TIME@all 0.303 -epoch: [275/350][20/50] time 0.297 (0.303) data 0.001 (0.012) eta 0:19:05 loss 1.1076 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.001 (0.006) eta 0:19:01 loss 1.1221 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.658, TIME@all 0.303 -epoch: [275/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:19:05 loss 1.0714 (1.0631) acc 100.0000 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:19:01 loss 1.0900 (1.0698) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.019, TIME@all 0.303 -epoch: [275/350][20/50] time 0.298 (0.303) data 0.000 (0.013) eta 0:19:05 loss 1.0585 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:19:01 loss 1.1035 (1.0742) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.780, TIME@all 0.303 -epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0670 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0632 (1.0715) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.284, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0757 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0584 (1.0701) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.208, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0678 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.1087 (1.0689) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.246, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:18:55 loss 1.0778 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:18:46 loss 1.0803 (1.0694) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.277, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.1130 (1.0629) acc 96.8750 (99.8438) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.2308 (1.0750) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 843.466, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.0488 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0567 (1.0705) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.415, TIME@all 0.304 -epoch: [276/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:18:55 loss 1.0651 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0903 (1.0672) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.252, TIME@all 0.304 -epoch: [276/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:18:55 loss 1.0478 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:18:46 loss 1.0819 (1.0697) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.591, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0512 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.1427 (1.0732) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 844.865, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0702 (1.0613) acc 100.0000 (99.8438) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.0508 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.903, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0585 (1.0581) acc 100.0000 (99.8438) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:18:29 loss 1.0751 (1.0658) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.021, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0540 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.007) eta 0:18:29 loss 1.0959 (1.0664) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 845.060, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0622 (1.0611) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.1115 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.927, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:18:35 loss 1.0681 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:18:29 loss 1.1070 (1.0657) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 844.845, TIME@all 0.303 -epoch: [277/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0711 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.303 (0.303) data 0.001 (0.006) eta 0:18:29 loss 1.0635 (1.0669) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.873, TIME@all 0.303 -epoch: [277/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:18:35 loss 1.0669 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.304 (0.303) data 0.000 (0.006) eta 0:18:29 loss 1.0856 (1.0698) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 845.275, TIME@all 0.303 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0979 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0538 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.959, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:18:25 loss 1.0705 (1.0520) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:18:18 loss 1.0525 (1.0660) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.871, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:18:25 loss 1.0723 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:18:18 loss 1.0634 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.889, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0879 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0549 (1.0646) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.897, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0757 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0818 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.905, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.014) eta 0:18:25 loss 1.0746 (1.0549) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0585 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.083, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.1031 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0545 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.028, TIME@all 0.304 -epoch: [278/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:18:25 loss 1.0924 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:18:18 loss 1.0620 (1.0662) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.244, TIME@all 0.304 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0683 (1.0519) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0861 (1.0631) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.834, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0710 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0912 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.891, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0641 (1.0512) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1315 (1.0630) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.917, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:18:07 loss 1.0997 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:18:01 loss 1.0756 (1.0665) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.040, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0874 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0880 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.827, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0630 (1.0519) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1170 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.839, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0904 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.1431 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.982, TIME@all 0.303 -epoch: [279/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:18:07 loss 1.0675 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:18:01 loss 1.0764 (1.0601) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.053, TIME@all 0.303 -epoch: [280/350][20/50] time 0.312 (0.307) data 0.001 (0.012) eta 0:18:02 loss 1.0533 (1.0573) acc 100.0000 (99.8438) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:17:50 loss 1.0461 (1.0629) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 840.540, TIME@all 0.305 -epoch: [280/350][20/50] time 0.297 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0650 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0533 (1.0646) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 839.562, TIME@all 0.305 -epoch: [280/350][20/50] time 0.295 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0594 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.006) eta 0:17:52 loss 1.0547 (1.0610) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.563, TIME@all 0.305 -epoch: [280/350][20/50] time 0.308 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.1720 (1.0624) acc 96.8750 (99.8438) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:52 loss 1.0489 (1.0695) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.499, TIME@all 0.305 -epoch: [280/350][20/50] time 0.309 (0.307) data 0.000 (0.013) eta 0:18:02 loss 1.0662 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:52 loss 1.0575 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 839.487, TIME@all 0.305 -epoch: [280/350][20/50] time 0.308 (0.307) data 0.000 (0.014) eta 0:18:02 loss 1.0684 (1.0595) acc 100.0000 (99.8438) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0538 (1.0701) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 839.673, TIME@all 0.305 -epoch: [280/350][20/50] time 0.309 (0.307) data 0.000 (0.014) eta 0:18:02 loss 1.0594 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.305) data 0.000 (0.007) eta 0:17:52 loss 1.0601 (1.0697) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 839.623, TIME@all 0.305 -epoch: [280/350][20/50] time 0.306 (0.308) data 0.000 (0.013) eta 0:18:06 loss 1.0705 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.304 (0.306) data 0.000 (0.007) eta 0:17:53 loss 1.0543 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 838.894, TIME@all 0.305 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:17:37 loss 1.0597 (1.0586) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0692 (1.0652) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.894, TIME@all 0.303 -epoch: [281/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0612 (1.0597) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0702 (1.0693) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.946, TIME@all 0.303 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.1250 (1.0636) acc 96.8750 (99.5312) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0493 (1.0690) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.992, TIME@all 0.303 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:17:37 loss 1.0574 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:17:30 loss 1.0405 (1.0675) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.120, TIME@all 0.303 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0519 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0672 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.053, TIME@all 0.303 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.1104 (1.0607) acc 96.8750 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0530 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.900, TIME@all 0.303 -epoch: [281/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0578 (1.0610) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0766 (1.0633) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.929, TIME@all 0.303 -epoch: [281/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:17:37 loss 1.0561 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [281/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:17:30 loss 1.0460 (1.0644) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.288, TIME@all 0.303 -epoch: [282/350][20/50] time 0.302 (0.303) data 0.001 (0.012) eta 0:17:20 loss 1.1048 (1.0598) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0745 (1.0647) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.669, TIME@all 0.304 -epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:17:20 loss 1.0925 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0690 (1.0648) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.679, TIME@all 0.304 -epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:17:20 loss 1.0776 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:17:17 loss 1.0476 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.748, TIME@all 0.304 -epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0966 (1.0564) acc 100.0000 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0523 (1.0660) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 842.883, TIME@all 0.304 -epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0591 (1.0523) acc 100.0000 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0458 (1.0612) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.675, TIME@all 0.304 -epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.1471 (1.0575) acc 96.8750 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0600 (1.0621) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.980, TIME@all 0.304 -epoch: [282/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.0932 (1.0593) acc 100.0000 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0532 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.694, TIME@all 0.304 -epoch: [282/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:17:20 loss 1.1286 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:17:17 loss 1.0808 (1.0665) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.821, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 0:17:11 loss 1.0579 (1.0543) acc 100.0000 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.1423 (1.0696) acc 96.8750 (99.6875) lr 0.002600 -FPS@all 843.036, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0686 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0746 (1.0681) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.129, TIME@all 0.304 -epoch: [283/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0552 (1.0503) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0790 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.070, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.011) eta 0:17:11 loss 1.0480 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0966 (1.0662) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.080, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.013) eta 0:17:11 loss 1.1076 (1.0576) acc 96.8750 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:17:02 loss 1.0765 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.280, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0530 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:17:02 loss 1.0583 (1.0672) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.096, TIME@all 0.304 -epoch: [283/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:17:12 loss 1.0511 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:17:02 loss 1.0594 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.397, TIME@all 0.304 -epoch: [283/350][20/50] time 0.303 (0.305) data 0.000 (0.012) eta 0:17:11 loss 1.0531 (1.0569) acc 100.0000 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:17:02 loss 1.0600 (1.0670) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.232, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1720 (1.0671) acc 96.8750 (99.6875) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2800 (1.0765) acc 93.7500 (99.5312) lr 0.002600 -FPS@all 843.314, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1454 (1.0575) acc 96.8750 (99.8438) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1541 (1.0726) acc 96.8750 (99.6094) lr 0.002600 -FPS@all 843.373, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.0751 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1162 (1.0679) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.340, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1062 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2198 (1.0676) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.348, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:16:55 loss 1.1267 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2419 (1.0765) acc 96.8750 (99.6094) lr 0.002600 -FPS@all 843.323, TIME@all 0.304 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:16:54 loss 1.0534 (1.0565) acc 100.0000 (99.8438) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:16:47 loss 1.2155 (1.0670) acc 93.7500 (99.6875) lr 0.002600 -FPS@all 843.530, TIME@all 0.303 -epoch: [284/350][20/50] time 0.306 (0.305) data 0.000 (0.013) eta 0:16:55 loss 1.1203 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.1152 (1.0699) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.474, TIME@all 0.304 -epoch: [284/350][20/50] time 0.302 (0.305) data 0.000 (0.012) eta 0:16:54 loss 1.0785 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:16:47 loss 1.2171 (1.0727) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 843.557, TIME@all 0.303 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:16:40 loss 1.2068 (1.0705) acc 96.8750 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.303 (0.304) data 0.001 (0.006) eta 0:16:32 loss 1.0658 (1.0721) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.262, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.012) eta 0:16:40 loss 1.1491 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:16:32 loss 1.0466 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.137, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.1622 (1.0627) acc 100.0000 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0568 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.199, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0785 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0520 (1.0635) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.179, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0962 (1.0617) acc 100.0000 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0559 (1.0697) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.185, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.1104 (1.0635) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0450 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.378, TIME@all 0.304 -epoch: [285/350][20/50] time 0.310 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0891 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0778 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.318, TIME@all 0.304 -epoch: [285/350][20/50] time 0.309 (0.305) data 0.000 (0.013) eta 0:16:40 loss 1.0589 (1.0600) acc 100.0000 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:16:32 loss 1.0484 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.591, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0980 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0531 (1.0631) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.188, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0686 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0481 (1.0616) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.265, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:16:22 loss 1.0631 (1.0501) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.001 (0.007) eta 0:16:16 loss 1.0571 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.417, TIME@all 0.304 -epoch: [286/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:16:23 loss 1.1103 (1.0581) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.001 (0.007) eta 0:16:17 loss 1.0535 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.194, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.1400 (1.0573) acc 96.8750 (99.8438) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0612 (1.0677) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.223, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.1348 (1.0567) acc 100.0000 (99.8438) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0547 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.235, TIME@all 0.304 -epoch: [286/350][20/50] time 0.300 (0.304) data 0.000 (0.012) eta 0:16:23 loss 1.0964 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:16:17 loss 1.0530 (1.0634) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.522, TIME@all 0.304 -epoch: [286/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:16:22 loss 1.0648 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:16:16 loss 1.0622 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.362, TIME@all 0.304 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.011) eta 0:16:05 loss 1.0551 (1.0577) acc 100.0000 (99.8438) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0667 (1.0677) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.969, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0588 (1.0518) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0522 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.002, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0813 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0642 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.037, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.1032 (1.0617) acc 96.8750 (99.8438) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0709 (1.0695) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.181, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0818 (1.0545) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0715 (1.0671) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.015, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0627 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.1223 (1.0679) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 843.986, TIME@all 0.303 -epoch: [287/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:16:05 loss 1.0684 (1.0578) acc 100.0000 (99.8438) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0523 (1.0666) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.125, TIME@all 0.303 -epoch: [287/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:16:05 loss 1.0539 (1.0520) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:16:00 loss 1.0563 (1.0606) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.293, TIME@all 0.303 -epoch: [288/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:15:51 loss 1.0736 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:15:44 loss 1.0518 (1.0669) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.549, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:15:51 loss 1.0477 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:15:44 loss 1.0719 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.480, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.1112 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0410 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.512, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:15:51 loss 1.0542 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:15:44 loss 1.0574 (1.0623) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.726, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0803 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0481 (1.0633) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.535, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:15:51 loss 1.0808 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0443 (1.0644) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.502, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0653 (1.0609) acc 100.0000 (99.8438) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0555 (1.0664) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.841, TIME@all 0.303 -epoch: [288/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:15:51 loss 1.0884 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:44 loss 1.0457 (1.0660) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.680, TIME@all 0.303 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:15:36 loss 1.1834 (1.0637) acc 96.8750 (99.8438) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.0840 (1.0735) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 842.896, TIME@all 0.304 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:15:36 loss 1.0858 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.1295 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.961, TIME@all 0.304 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0950 (1.0606) acc 100.0000 (99.8438) lr 0.002600 -epoch: [289/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:15:30 loss 1.0636 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.899, TIME@all 0.304 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0959 (1.0568) acc 96.8750 (99.8438) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.0621 (1.0660) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.943, TIME@all 0.304 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.0726 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:15:30 loss 1.0860 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.916, TIME@all 0.304 -epoch: [289/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:15:35 loss 1.0827 (1.0518) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.1400 (1.0691) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 843.115, TIME@all 0.304 -epoch: [289/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:15:36 loss 1.1103 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.0478 (1.0650) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.258, TIME@all 0.304 -epoch: [289/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:15:35 loss 1.1344 (1.0572) acc 96.8750 (99.8438) lr 0.002600 -epoch: [289/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:15:30 loss 1.1097 (1.0656) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.094, TIME@all 0.304 -epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0661 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0539 (1.0649) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.182, TIME@all 0.304 -epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0466 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0443 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.221, TIME@all 0.304 -epoch: [290/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0681 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0660 (1.0647) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.206, TIME@all 0.304 -epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0706 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:15 loss 1.0493 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.202, TIME@all 0.304 -epoch: [290/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0420 (1.0532) acc 100.0000 (99.8438) lr 0.002600 -epoch: [290/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0488 (1.0671) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.562, TIME@all 0.303 -epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0651 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:14 loss 1.0478 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.371, TIME@all 0.304 -epoch: [290/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:15:22 loss 1.0685 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:15:15 loss 1.0675 (1.0707) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.176, TIME@all 0.304 -epoch: [290/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:15:22 loss 1.0498 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:15:14 loss 1.0552 (1.0687) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.355, TIME@all 0.304 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.0687 (1.0522) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.1356 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.735, TIME@all 0.303 -epoch: [291/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.1037 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.1135 (1.0667) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.830, TIME@all 0.303 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:15:06 loss 1.0501 (1.0567) acc 100.0000 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.006) eta 0:14:58 loss 1.0736 (1.0657) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.746, TIME@all 0.303 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.0569 (1.0584) acc 100.0000 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.1279 (1.0713) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.777, TIME@all 0.303 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.1646 (1.0621) acc 96.8750 (99.6875) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0843 (1.0733) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 843.762, TIME@all 0.303 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:15:06 loss 1.0796 (1.0667) acc 100.0000 (99.6875) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0588 (1.0677) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.949, TIME@all 0.303 -epoch: [291/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.1336 (1.0576) acc 96.8750 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.299 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0545 (1.0617) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.087, TIME@all 0.303 -epoch: [291/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:15:06 loss 1.0547 (1.0515) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:14:58 loss 1.0941 (1.0653) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.900, TIME@all 0.303 -epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.1118 (1.0599) acc 100.0000 (99.8438) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:44 loss 1.0494 (1.0686) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.326, TIME@all 0.304 -epoch: [292/350][20/50] time 0.304 (0.304) data 0.001 (0.012) eta 0:14:50 loss 1.0617 (1.0514) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0601 (1.0659) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.253, TIME@all 0.304 -epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0708 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0668 (1.0718) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.311, TIME@all 0.304 -epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0704 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0709 (1.0728) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.477, TIME@all 0.304 -epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0619 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0592 (1.0717) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.275, TIME@all 0.304 -epoch: [292/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0748 (1.0507) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:14:44 loss 1.0420 (1.0654) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.284, TIME@all 0.304 -epoch: [292/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:14:50 loss 1.0749 (1.0578) acc 100.0000 (99.8438) lr 0.002600 -epoch: [292/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:14:44 loss 1.0517 (1.0700) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.428, TIME@all 0.304 -epoch: [292/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:14:50 loss 1.0821 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:14:45 loss 1.0470 (1.0689) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.635, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0451 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0497 (1.0643) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.265, TIME@all 0.303 -epoch: [293/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0656 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0867 (1.0694) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.103, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.1205 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0840 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 845.207, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0546 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0547 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.146, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.1232 (1.0598) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0554 (1.0639) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 845.168, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0474 (1.0603) acc 100.0000 (99.8438) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.1093 (1.0679) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 845.358, TIME@all 0.303 -epoch: [293/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0545 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0809 (1.0694) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.291, TIME@all 0.303 -epoch: [293/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:14:33 loss 1.0629 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:14:27 loss 1.0621 (1.0686) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 845.392, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:14:17 loss 1.0395 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:14:14 loss 1.0578 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.909, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0541 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0542 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 843.777, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:17 loss 1.0671 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0737 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.789, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.014) eta 0:14:17 loss 1.0531 (1.0523) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0612 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.979, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0566 (1.0512) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0678 (1.0600) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.834, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:18 loss 1.0731 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0563 (1.0653) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.787, TIME@all 0.303 -epoch: [294/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:14:17 loss 1.0514 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0775 (1.0661) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.132, TIME@all 0.303 -epoch: [294/350][20/50] time 0.303 (0.303) data 0.001 (0.013) eta 0:14:17 loss 1.0552 (1.0546) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:14:14 loss 1.0593 (1.0627) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.945, TIME@all 0.303 -epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.012) eta 0:14:01 loss 1.0501 (1.0563) acc 100.0000 (99.8438) lr 0.002600 -epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:13:57 loss 1.0683 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.721, TIME@all 0.303 -epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:14:01 loss 1.0579 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:13:57 loss 1.1048 (1.0636) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.665, TIME@all 0.303 -epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0479 (1.0588) acc 100.0000 (99.6875) lr 0.002600 -epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0863 (1.0691) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.695, TIME@all 0.303 -epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0584 (1.0513) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0929 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.676, TIME@all 0.303 -epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0456 (1.0518) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1197 (1.0620) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.859, TIME@all 0.303 -epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0485 (1.0575) acc 100.0000 (99.8438) lr 0.002600 -epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1007 (1.0711) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.667, TIME@all 0.303 -epoch: [295/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0518 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.0784 (1.0656) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.821, TIME@all 0.303 -epoch: [295/350][20/50] time 0.299 (0.303) data 0.000 (0.013) eta 0:14:01 loss 1.0674 (1.0615) acc 100.0000 (99.8438) lr 0.002600 -epoch: [295/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:13:57 loss 1.1069 (1.0724) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 845.137, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1634 (1.0658) acc 96.8750 (99.5312) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0602 (1.0664) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.215, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1086 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0413 (1.0665) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.151, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:13:48 loss 1.2177 (1.0616) acc 96.8750 (99.6875) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:13:43 loss 1.1168 (1.0727) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 844.318, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1431 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0646 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.130, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:48 loss 1.1645 (1.0683) acc 96.8750 (99.6875) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:13:43 loss 1.0715 (1.0737) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 844.379, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1463 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0698 (1.0731) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.189, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1283 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0569 (1.0633) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.179, TIME@all 0.303 -epoch: [296/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:49 loss 1.1528 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [296/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:13:43 loss 1.0539 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.510, TIME@all 0.303 -epoch: [297/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0772 (1.0523) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.1012 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.001, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0672 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0883 (1.0670) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 843.004, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0754 (1.0550) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0877 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.033, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0859 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.1186 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.975, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:34 loss 1.0593 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:13:29 loss 1.0604 (1.0689) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 843.172, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0704 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0894 (1.0633) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 843.121, TIME@all 0.304 -epoch: [297/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0606 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:29 loss 1.0660 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.964, TIME@all 0.304 -epoch: [297/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:35 loss 1.0706 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:13:29 loss 1.0829 (1.0661) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 843.313, TIME@all 0.304 -epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0506 (1.0529) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.1675 (1.0723) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.243, TIME@all 0.303 -epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.1373 (1.0630) acc 96.8750 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.0736 (1.0727) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.249, TIME@all 0.303 -epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0638 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1242 (1.0705) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.277, TIME@all 0.303 -epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0563 (1.0596) acc 100.0000 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:13:13 loss 1.1070 (1.0652) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.273, TIME@all 0.303 -epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0493 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1960 (1.0693) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 844.460, TIME@all 0.303 -epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:13:19 loss 1.0650 (1.0590) acc 100.0000 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:13:13 loss 1.0687 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.607, TIME@all 0.303 -epoch: [298/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0585 (1.0627) acc 100.0000 (99.6875) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1180 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.263, TIME@all 0.303 -epoch: [298/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:13:19 loss 1.0559 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:13:13 loss 1.1101 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.420, TIME@all 0.303 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.011) eta 0:13:04 loss 1.0741 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.1183 (1.0694) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.189, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.1197 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.0744 (1.0753) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.278, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.1092 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.0690 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 842.231, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:04 loss 1.1008 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.309 (0.304) data 0.000 (0.007) eta 0:12:59 loss 1.0770 (1.0682) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.204, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:03 loss 1.0986 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.308 (0.304) data 0.001 (0.007) eta 0:12:59 loss 1.0978 (1.0693) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 842.418, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:13:04 loss 1.0732 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.309 (0.304) data 0.001 (0.006) eta 0:12:59 loss 1.0619 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 842.187, TIME@all 0.304 -epoch: [299/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:13:03 loss 1.0703 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:12:59 loss 1.1044 (1.0729) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 842.563, TIME@all 0.304 -epoch: [299/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:13:04 loss 1.0924 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:12:59 loss 1.0985 (1.0738) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 842.367, TIME@all 0.304 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0527 (1.0507) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0427 (1.0642) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.477, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:12:48 loss 1.0526 (1.0532) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:12:43 loss 1.0642 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.408, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:12:48 loss 1.0585 (1.0581) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:12:43 loss 1.0474 (1.0692) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.487, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:12:48 loss 1.0780 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0637 (1.0670) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 844.580, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0501 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0477 (1.0695) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.406, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0724 (1.0539) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:43 loss 1.0564 (1.0699) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 844.408, TIME@all 0.303 -epoch: [300/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0697 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0597 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 844.557, TIME@all 0.303 -epoch: [300/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:12:48 loss 1.0555 (1.0539) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:12:42 loss 1.0661 (1.0719) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 844.767, TIME@all 0.303 -epoch: [301/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0613 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0688 (1.0622) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.063, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0478 (1.0518) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0621 (1.0612) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 845.087, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0659 (1.0536) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0626 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.983, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:12:32 loss 1.0816 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.306 (0.303) data 0.001 (0.007) eta 0:12:26 loss 1.0588 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 845.201, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0474 (1.0500) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.306 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0862 (1.0618) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.037, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.012) eta 0:12:32 loss 1.0599 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.305 (0.303) data 0.001 (0.006) eta 0:12:26 loss 1.0718 (1.0663) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.026, TIME@all 0.303 -epoch: [301/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:12:32 loss 1.0671 (1.0585) acc 100.0000 (99.8438) lr 0.000260 -epoch: [301/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:12:26 loss 1.0885 (1.0695) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 845.159, TIME@all 0.303 -epoch: [301/350][20/50] time 0.301 (0.303) data 0.001 (0.012) eta 0:12:32 loss 1.0705 (1.0556) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.305 (0.303) data 0.000 (0.006) eta 0:12:26 loss 1.0630 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 845.315, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.011) eta 0:12:17 loss 1.0545 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.1234 (1.0642) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 843.593, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0607 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0692 (1.0625) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.664, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0515 (1.0529) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:12:12 loss 1.0702 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.804, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0560 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0612 (1.0671) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.606, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0507 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0929 (1.0646) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.612, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:12:17 loss 1.0614 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0843 (1.0630) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.629, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:12:17 loss 1.0459 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0934 (1.0620) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.759, TIME@all 0.303 -epoch: [302/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:12:17 loss 1.0646 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:12:12 loss 1.0833 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.748, TIME@all 0.303 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0552 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1023 (1.0691) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.355, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.011) eta 0:12:03 loss 1.0817 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0673 (1.0661) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.273, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.1474 (1.0563) acc 96.8750 (99.6875) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0567 (1.0650) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 842.366, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0795 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0559 (1.0672) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.299, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0639 (1.0606) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1074 (1.0709) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.327, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:12:03 loss 1.0652 (1.0569) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.0723 (1.0649) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.676, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:12:03 loss 1.0798 (1.0552) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:11:58 loss 1.0554 (1.0693) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.481, TIME@all 0.304 -epoch: [303/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:12:03 loss 1.0756 (1.0622) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:11:58 loss 1.1654 (1.0738) acc 96.8750 (99.7656) lr 0.000260 -FPS@all 842.447, TIME@all 0.304 -epoch: [304/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0579 (1.0548) acc 100.0000 (99.8438) lr 0.000260 -epoch: [304/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:11:41 loss 1.1258 (1.0630) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.293, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0555 (1.0578) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0689 (1.0661) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.237, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:11:49 loss 1.0682 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:41 loss 1.0641 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.179, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:11:48 loss 1.0505 (1.0595) acc 100.0000 (99.8438) lr 0.000260 -epoch: [304/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:11:40 loss 1.1018 (1.0694) acc 96.8750 (99.6875) lr 0.000260 -FPS@all 845.371, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0590 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.301 (0.303) data 0.001 (0.007) eta 0:11:41 loss 1.1090 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 845.564, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0491 (1.0556) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0495 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.173, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:49 loss 1.0767 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:11:41 loss 1.0829 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.204, TIME@all 0.303 -epoch: [304/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:11:48 loss 1.0511 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:11:40 loss 1.0836 (1.0622) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.329, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.011) eta 0:11:32 loss 1.0647 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0528 (1.0639) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.663, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:11:32 loss 1.0748 (1.0575) acc 100.0000 (99.8438) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0769 (1.0652) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.707, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:11:32 loss 1.1233 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0491 (1.0658) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.744, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0719 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0806 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.694, TIME@all 0.303 -epoch: [305/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:32 loss 1.1035 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0646 (1.0647) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.996, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0791 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0515 (1.0709) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.832, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:11:32 loss 1.1262 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:26 loss 1.0715 (1.0681) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.890, TIME@all 0.303 -epoch: [305/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:11:32 loss 1.0658 (1.0550) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:26 loss 1.0539 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.689, TIME@all 0.303 -epoch: [306/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:11:17 loss 1.1035 (1.0589) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1163 (1.0724) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.654, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:17 loss 1.1553 (1.0595) acc 96.8750 (99.8438) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1108 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.686, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:17 loss 1.1215 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1183 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.707, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:17 loss 1.1302 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:11 loss 1.0618 (1.0696) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.859, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:11:17 loss 1.0814 (1.0606) acc 100.0000 (99.8438) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:11:11 loss 1.0912 (1.0700) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.653, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.001 (0.012) eta 0:11:17 loss 1.1069 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.1218 (1.0674) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.636, TIME@all 0.303 -epoch: [306/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:17 loss 1.1202 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.0693 (1.0649) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.772, TIME@all 0.303 -epoch: [306/350][20/50] time 0.303 (0.304) data 0.001 (0.012) eta 0:11:17 loss 1.0792 (1.0570) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:11:11 loss 1.0767 (1.0636) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.038, TIME@all 0.303 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.1065 (1.0654) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0494 (1.0676) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.331, TIME@all 0.304 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.0613 (1.0666) acc 100.0000 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0532 (1.0717) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.388, TIME@all 0.304 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0517 (1.0605) acc 100.0000 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0524 (1.0745) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 843.375, TIME@all 0.304 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0660 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0617 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.338, TIME@all 0.304 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.1055 (1.0618) acc 96.8750 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0463 (1.0715) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.513, TIME@all 0.303 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0868 (1.0643) acc 100.0000 (99.6875) lr 0.000260 -epoch: [307/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:10:56 loss 1.0612 (1.0657) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.485, TIME@all 0.304 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:11:03 loss 1.0701 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:10:56 loss 1.0967 (1.0703) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.715, TIME@all 0.303 -epoch: [307/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:11:03 loss 1.0865 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:10:56 loss 1.0514 (1.0697) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.338, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:10:48 loss 1.0724 (1.0556) acc 100.0000 (99.8438) lr 0.000260 -epoch: [308/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0534 (1.0604) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.454, TIME@all 0.304 -epoch: [308/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:10:48 loss 1.0874 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0491 (1.0650) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.396, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.014) eta 0:10:48 loss 1.0871 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0535 (1.0628) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.427, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:10:48 loss 1.0875 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0440 (1.0613) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.386, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.0797 (1.0572) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0484 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.418, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.1035 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0409 (1.0667) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.604, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.1178 (1.0629) acc 96.8750 (99.8438) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0452 (1.0691) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.768, TIME@all 0.304 -epoch: [308/350][20/50] time 0.305 (0.305) data 0.001 (0.014) eta 0:10:48 loss 1.0775 (1.0499) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:10:42 loss 1.0857 (1.0670) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.562, TIME@all 0.304 -epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0697 (1.0597) acc 100.0000 (99.8438) lr 0.000260 -epoch: [309/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.1301 (1.0670) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 843.768, TIME@all 0.303 -epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0584 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:10:25 loss 1.0552 (1.0639) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.826, TIME@all 0.303 -epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0471 (1.0584) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0546 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.799, TIME@all 0.303 -epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.0664 (1.0650) acc 100.0000 (99.6875) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0533 (1.0679) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.997, TIME@all 0.303 -epoch: [309/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0701 (1.0587) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0719 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.792, TIME@all 0.303 -epoch: [309/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:31 loss 1.0475 (1.0573) acc 100.0000 (99.8438) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0943 (1.0649) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.177, TIME@all 0.303 -epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.0660 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0610 (1.0621) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.947, TIME@all 0.303 -epoch: [309/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:10:31 loss 1.1896 (1.0657) acc 96.8750 (99.8438) lr 0.000260 -epoch: [309/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:10:25 loss 1.0939 (1.0695) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.799, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0579 (1.0514) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0590 (1.0615) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.754, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0616 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.299 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0726 (1.0703) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.779, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0741 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.1570 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.910, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0885 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.300 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0719 (1.0646) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.777, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0433 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.001 (0.006) eta 0:10:09 loss 1.0671 (1.0659) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.772, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0764 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:10:09 loss 1.0769 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.971, TIME@all 0.303 -epoch: [310/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:10:16 loss 1.0544 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.007) eta 0:10:09 loss 1.0598 (1.0656) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 844.928, TIME@all 0.303 -epoch: [310/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:10:17 loss 1.0544 (1.0498) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.301 (0.303) data 0.000 (0.006) eta 0:10:09 loss 1.0494 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.145, TIME@all 0.303 -epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0659 (1.0529) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0601 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.895, TIME@all 0.303 -epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0603 (1.0511) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0703 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.930, TIME@all 0.303 -epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:10:01 loss 1.0614 (1.0500) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.006) eta 0:09:54 loss 1.0457 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.880, TIME@all 0.303 -epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0711 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0582 (1.0593) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.986, TIME@all 0.303 -epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0629 (1.0493) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0653 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.909, TIME@all 0.303 -epoch: [311/350][20/50] time 0.302 (0.304) data 0.001 (0.014) eta 0:10:01 loss 1.0621 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0477 (1.0630) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 845.086, TIME@all 0.303 -epoch: [311/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0747 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0570 (1.0631) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.039, TIME@all 0.303 -epoch: [311/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:10:01 loss 1.0526 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.305 (0.303) data 0.000 (0.007) eta 0:09:54 loss 1.0625 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 845.226, TIME@all 0.303 -epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0605 (1.0649) acc 100.0000 (99.8438) lr 0.000260 -epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.0668 (1.0677) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.749, TIME@all 0.303 -epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.012) eta 0:09:45 loss 1.0585 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:09:40 loss 1.0531 (1.0649) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.651, TIME@all 0.303 -epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0734 (1.0605) acc 100.0000 (99.8438) lr 0.000260 -epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1035 (1.0662) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.637, TIME@all 0.303 -epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.014) eta 0:09:45 loss 1.0490 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1028 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.859, TIME@all 0.303 -epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0521 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1479 (1.0644) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 844.062, TIME@all 0.303 -epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0524 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1189 (1.0651) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.674, TIME@all 0.303 -epoch: [312/350][20/50] time 0.300 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0502 (1.0618) acc 100.0000 (99.6875) lr 0.000260 -epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.1560 (1.0675) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.833, TIME@all 0.303 -epoch: [312/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:09:45 loss 1.0651 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:09:40 loss 1.0988 (1.0624) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.699, TIME@all 0.303 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:09:31 loss 1.0488 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0566 (1.0684) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.485, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0550 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0679 (1.0642) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.533, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0600 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0884 (1.0685) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.580, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0687 (1.0632) acc 100.0000 (99.8438) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0579 (1.0685) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.485, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:09:31 loss 1.1140 (1.0663) acc 96.8750 (99.6875) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0846 (1.0742) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 842.506, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:09:31 loss 1.0889 (1.0546) acc 100.0000 (99.8438) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:09:25 loss 1.0824 (1.0662) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.696, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:09:31 loss 1.0480 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:09:25 loss 1.0472 (1.0658) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.647, TIME@all 0.304 -epoch: [313/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:09:31 loss 1.0686 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:25 loss 1.0690 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.839, TIME@all 0.304 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0552 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.0599 (1.0618) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.943, TIME@all 0.303 -epoch: [314/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0494 (1.0533) acc 100.0000 (99.8438) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.0736 (1.0605) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.947, TIME@all 0.303 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:09:15 loss 1.0923 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:09:09 loss 1.1173 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.995, TIME@all 0.303 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0633 (1.0504) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0672 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.941, TIME@all 0.303 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0554 (1.0549) acc 100.0000 (99.8438) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.1298 (1.0723) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 843.961, TIME@all 0.303 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0446 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0604 (1.0651) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.115, TIME@all 0.303 -epoch: [314/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0595 (1.0510) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0693 (1.0611) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.079, TIME@all 0.303 -epoch: [314/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:09:15 loss 1.0495 (1.0494) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:09:09 loss 1.0498 (1.0642) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.202, TIME@all 0.303 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:09:01 loss 1.0857 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1019 (1.0668) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.405, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:09:01 loss 1.1000 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1180 (1.0671) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.406, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0835 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0897 (1.0656) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.390, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0636 (1.0593) acc 100.0000 (99.8438) lr 0.000260 -epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0980 (1.0643) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.459, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:09:01 loss 1.0724 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:55 loss 1.0716 (1.0691) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.550, TIME@all 0.304 -epoch: [315/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0779 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0567 (1.0645) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.678, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.1024 (1.0581) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.0769 (1.0648) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.546, TIME@all 0.304 -epoch: [315/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:09:01 loss 1.0836 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:55 loss 1.1057 (1.0655) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.397, TIME@all 0.304 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:08:47 loss 1.0600 (1.0558) acc 100.0000 (99.8438) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0691 (1.0668) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 844.045, TIME@all 0.303 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:08:47 loss 1.0526 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0479 (1.0616) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.083, TIME@all 0.303 -epoch: [316/350][20/50] time 0.306 (0.305) data 0.000 (0.012) eta 0:08:47 loss 1.0455 (1.0568) acc 100.0000 (99.8438) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0438 (1.0629) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.058, TIME@all 0.303 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:46 loss 1.0539 (1.0505) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:08:39 loss 1.0596 (1.0621) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.241, TIME@all 0.303 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:08:47 loss 1.0546 (1.0516) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0551 (1.0619) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.029, TIME@all 0.303 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:46 loss 1.0428 (1.0526) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0725 (1.0634) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.188, TIME@all 0.303 -epoch: [316/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:08:46 loss 1.0473 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.298 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0807 (1.0635) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.475, TIME@all 0.303 -epoch: [316/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:08:46 loss 1.0510 (1.0511) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:08:39 loss 1.0587 (1.0636) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.039, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:08:29 loss 1.0677 (1.0600) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:24 loss 1.0456 (1.0649) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.523, TIME@all 0.303 -epoch: [317/350][20/50] time 0.300 (0.303) data 0.001 (0.013) eta 0:08:29 loss 1.0457 (1.0645) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0897 (1.0719) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.584, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0451 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0647 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.542, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0547 (1.0558) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0757 (1.0666) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.585, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0528 (1.0637) acc 100.0000 (99.6875) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0925 (1.0737) acc 96.8750 (99.6875) lr 0.000260 -FPS@all 843.551, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:29 loss 1.0488 (1.0599) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.0697 (1.0699) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.702, TIME@all 0.303 -epoch: [317/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:08:29 loss 1.0571 (1.0626) acc 100.0000 (99.8438) lr 0.000260 -epoch: [317/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:08:24 loss 1.0593 (1.0670) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.928, TIME@all 0.303 -epoch: [317/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:29 loss 1.0660 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:24 loss 1.1176 (1.0682) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.723, TIME@all 0.303 -epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:08:14 loss 1.0630 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0581 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.926, TIME@all 0.303 -epoch: [318/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:08:14 loss 1.0654 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0669 (1.0658) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.965, TIME@all 0.303 -epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.0662 (1.0602) acc 100.0000 (99.8438) lr 0.000260 -epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0562 (1.0668) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.928, TIME@all 0.303 -epoch: [318/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.0802 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:08:09 loss 1.0619 (1.0632) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.920, TIME@all 0.303 -epoch: [318/350][20/50] time 0.301 (0.303) data 0.000 (0.014) eta 0:08:14 loss 1.0623 (1.0521) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0575 (1.0615) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.088, TIME@all 0.303 -epoch: [318/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:08:14 loss 1.0621 (1.0584) acc 100.0000 (99.8438) lr 0.000260 -epoch: [318/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:08:09 loss 1.0472 (1.0664) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.962, TIME@all 0.303 -epoch: [318/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:08:14 loss 1.0682 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0487 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.036, TIME@all 0.303 -epoch: [318/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:08:14 loss 1.1093 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:08:09 loss 1.0613 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.224, TIME@all 0.303 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0653 (1.0554) acc 100.0000 (99.8438) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0478 (1.0607) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.400, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.1132 (1.0622) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0458 (1.0656) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.380, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0745 (1.0577) acc 100.0000 (99.8438) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0461 (1.0570) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.465, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:08:01 loss 1.0570 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:07:54 loss 1.0536 (1.0636) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.379, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0783 (1.0594) acc 100.0000 (99.8438) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:07:54 loss 1.0458 (1.0612) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.548, TIME@all 0.304 -epoch: [319/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.1104 (1.0559) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:07:54 loss 1.0987 (1.0643) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.627, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0441 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:07:54 loss 1.0558 (1.0598) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.391, TIME@all 0.304 -epoch: [319/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:08:01 loss 1.0625 (1.0570) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.303 (0.304) data 0.001 (0.007) eta 0:07:54 loss 1.1173 (1.0615) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.565, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0734 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0504 (1.0615) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.822, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0989 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0476 (1.0637) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.879, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0758 (1.0607) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0511 (1.0700) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.873, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:07:46 loss 1.0749 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0874 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.909, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1238 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.001 (0.007) eta 0:07:39 loss 1.0557 (1.0631) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.858, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:07:46 loss 1.0995 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:39 loss 1.0771 (1.0673) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.023, TIME@all 0.304 -epoch: [320/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1131 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:39 loss 1.0534 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.987, TIME@all 0.304 -epoch: [320/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:07:46 loss 1.1228 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:39 loss 1.0518 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.154, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.0596 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.0589 (1.0609) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.849, TIME@all 0.304 -epoch: [321/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.1320 (1.0592) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.0705 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.904, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.1177 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.1212 (1.0639) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.906, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.1236 (1.0612) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0850 (1.0636) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.865, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.1159 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.1686 (1.0693) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.067, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.0807 (1.0623) acc 100.0000 (99.8438) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0941 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.017, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:07:29 loss 1.0614 (1.0623) acc 100.0000 (99.6875) lr 0.000260 -epoch: [321/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:07:24 loss 1.1331 (1.0703) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.891, TIME@all 0.304 -epoch: [321/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:07:29 loss 1.0667 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:07:24 loss 1.0720 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.232, TIME@all 0.304 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0466 (1.0593) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0938 (1.0643) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.352, TIME@all 0.303 -epoch: [322/350][20/50] time 0.303 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0576 (1.0621) acc 100.0000 (99.6875) lr 0.000260 -epoch: [322/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0711 (1.0680) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.357, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.011) eta 0:07:14 loss 1.0546 (1.0616) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.1261 (1.0736) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 844.370, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0932 (1.0596) acc 96.8750 (99.6875) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.1449 (1.0693) acc 96.8750 (99.7656) lr 0.000260 -FPS@all 844.320, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0709 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0823 (1.0664) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.319, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:07:14 loss 1.0678 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:07:07 loss 1.0947 (1.0688) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.508, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0520 (1.0669) acc 100.0000 (99.5312) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:07:08 loss 1.0922 (1.0760) acc 100.0000 (99.5312) lr 0.000260 -FPS@all 844.457, TIME@all 0.303 -epoch: [322/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:07:14 loss 1.0623 (1.0616) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.301 (0.304) data 0.001 (0.006) eta 0:07:08 loss 1.0741 (1.0681) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.703, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:06:59 loss 1.0495 (1.0516) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:06:53 loss 1.0828 (1.0657) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.767, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0682 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0820 (1.0632) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.786, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:06:59 loss 1.0582 (1.0509) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:06:53 loss 1.0698 (1.0666) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.787, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0487 (1.0519) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0845 (1.0697) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.752, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:06:59 loss 1.0443 (1.0507) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:06:53 loss 1.1086 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.772, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0523 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0666 (1.0634) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.967, TIME@all 0.303 -epoch: [323/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0463 (1.0519) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0758 (1.0685) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.062, TIME@all 0.303 -epoch: [323/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:06:59 loss 1.0591 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:53 loss 1.0836 (1.0638) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.899, TIME@all 0.303 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0540 (1.0524) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0483 (1.0592) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.213, TIME@all 0.304 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0531 (1.0519) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0632 (1.0589) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.241, TIME@all 0.304 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0685 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0642 (1.0623) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.241, TIME@all 0.304 -epoch: [324/350][20/50] time 0.310 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0507 (1.0484) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0440 (1.0598) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.282, TIME@all 0.304 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0465 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.309 (0.304) data 0.001 (0.006) eta 0:06:38 loss 1.0483 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.330, TIME@all 0.304 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0929 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:38 loss 1.1219 (1.0636) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 843.437, TIME@all 0.304 -epoch: [324/350][20/50] time 0.311 (0.304) data 0.000 (0.013) eta 0:06:44 loss 1.0612 (1.0668) acc 100.0000 (99.6875) lr 0.000260 -epoch: [324/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:38 loss 1.0863 (1.0670) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.391, TIME@all 0.304 -epoch: [324/350][20/50] time 0.312 (0.304) data 0.000 (0.012) eta 0:06:44 loss 1.0554 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:06:38 loss 1.0792 (1.0607) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.556, TIME@all 0.303 -epoch: [325/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0690 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1474 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.539, TIME@all 0.303 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0489 (1.0617) acc 100.0000 (99.8438) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:23 loss 1.0740 (1.0671) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.542, TIME@all 0.303 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0643 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:23 loss 1.0659 (1.0610) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.458, TIME@all 0.304 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:29 loss 1.0631 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1672 (1.0666) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.504, TIME@all 0.303 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0882 (1.0617) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:06:23 loss 1.1385 (1.0694) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.489, TIME@all 0.304 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0677 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:06:23 loss 1.0898 (1.0669) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.840, TIME@all 0.303 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0642 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:06:23 loss 1.1432 (1.0681) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.680, TIME@all 0.303 -epoch: [325/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:06:29 loss 1.0521 (1.0578) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:23 loss 1.1264 (1.0691) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.634, TIME@all 0.303 -epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0530 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0607 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.555, TIME@all 0.304 -epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0613 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0945 (1.0612) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.632, TIME@all 0.304 -epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0518 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0510 (1.0614) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.590, TIME@all 0.304 -epoch: [326/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:06:14 loss 1.0540 (1.0519) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0854 (1.0623) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.543, TIME@all 0.304 -epoch: [326/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0475 (1.0575) acc 100.0000 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0582 (1.0664) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.749, TIME@all 0.304 -epoch: [326/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0503 (1.0581) acc 100.0000 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:06:08 loss 1.0564 (1.0708) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.551, TIME@all 0.304 -epoch: [326/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:06:14 loss 1.0778 (1.0552) acc 96.8750 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:06:08 loss 1.0461 (1.0621) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.071, TIME@all 0.304 -epoch: [326/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:06:14 loss 1.0547 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:06:08 loss 1.0562 (1.0615) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.879, TIME@all 0.304 -epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.1138 (1.0587) acc 100.0000 (99.8438) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0651 (1.0683) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.459, TIME@all 0.304 -epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0532 (1.0511) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0551 (1.0617) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.374, TIME@all 0.304 -epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0599 (1.0569) acc 100.0000 (99.8438) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.1292 (1.0676) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.420, TIME@all 0.304 -epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:05:58 loss 1.0559 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:53 loss 1.0560 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.607, TIME@all 0.304 -epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0527 (1.0534) acc 100.0000 (99.8438) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0606 (1.0614) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.415, TIME@all 0.304 -epoch: [327/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0576 (1.0552) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0929 (1.0654) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.412, TIME@all 0.304 -epoch: [327/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0679 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0577 (1.0695) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.796, TIME@all 0.304 -epoch: [327/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:05:58 loss 1.0693 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:53 loss 1.0508 (1.0630) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.545, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0722 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0696 (1.0678) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.089, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:05:44 loss 1.1244 (1.0637) acc 100.0000 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:37 loss 1.1079 (1.0711) acc 96.8750 (99.7656) lr 0.000260 -FPS@all 843.065, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.011) eta 0:05:44 loss 1.0770 (1.0592) acc 100.0000 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.1262 (1.0718) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.040, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.001 (0.013) eta 0:05:44 loss 1.1897 (1.0645) acc 96.8750 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:05:37 loss 1.0699 (1.0697) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.274, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0615 (1.0552) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0946 (1.0638) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.062, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0693 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.1054 (1.0704) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.066, TIME@all 0.304 -epoch: [328/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:05:44 loss 1.1414 (1.0617) acc 96.8750 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:37 loss 1.0959 (1.0677) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.210, TIME@all 0.304 -epoch: [328/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:05:44 loss 1.0559 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:05:37 loss 1.0846 (1.0767) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 843.380, TIME@all 0.304 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.012) eta 0:05:27 loss 1.0837 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.006) eta 0:05:21 loss 1.0664 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.973, TIME@all 0.303 -epoch: [329/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:05:27 loss 1.0589 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.006) eta 0:05:21 loss 1.1538 (1.0637) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 844.919, TIME@all 0.303 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0547 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0489 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.932, TIME@all 0.303 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0990 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0696 (1.0682) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.938, TIME@all 0.303 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0717 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0645 (1.0630) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.933, TIME@all 0.303 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0705 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0504 (1.0618) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 845.093, TIME@all 0.303 -epoch: [329/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0817 (1.0714) acc 100.0000 (99.8438) lr 0.000260 -epoch: [329/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0654 (1.0720) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.333, TIME@all 0.303 -epoch: [329/350][20/50] time 0.304 (0.303) data 0.000 (0.013) eta 0:05:27 loss 1.0558 (1.0500) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.291 (0.303) data 0.000 (0.007) eta 0:05:21 loss 1.0495 (1.0604) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 845.154, TIME@all 0.303 -epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0555 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0550 (1.0643) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.964, TIME@all 0.304 -epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:05:12 loss 1.0755 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:07 loss 1.0459 (1.0623) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.027, TIME@all 0.304 -epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0623 (1.0565) acc 100.0000 (99.8438) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0449 (1.0649) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.960, TIME@all 0.304 -epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.012) eta 0:05:12 loss 1.0473 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:05:07 loss 1.0452 (1.0731) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.936, TIME@all 0.304 -epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0474 (1.0549) acc 100.0000 (99.8438) lr 0.000260 -epoch: [330/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:05:06 loss 1.0792 (1.0635) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.146, TIME@all 0.304 -epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0702 (1.0571) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0625 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.111, TIME@all 0.304 -epoch: [330/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0610 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.306 (0.304) data 0.001 (0.007) eta 0:05:07 loss 1.0522 (1.0659) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.945, TIME@all 0.304 -epoch: [330/350][20/50] time 0.301 (0.303) data 0.000 (0.013) eta 0:05:12 loss 1.0505 (1.0593) acc 100.0000 (99.6875) lr 0.000260 -epoch: [330/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:05:07 loss 1.0500 (1.0647) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.146, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0656 (1.0618) acc 100.0000 (99.8438) lr 0.000260 -epoch: [331/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0512 (1.0697) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.456, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0623 (1.0559) acc 100.0000 (99.8438) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0550 (1.0651) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.557, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0714 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0574 (1.0661) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.693, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0511 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0575 (1.0674) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.464, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:58 loss 1.0566 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:04:52 loss 1.0733 (1.0634) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.658, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:58 loss 1.0960 (1.0601) acc 96.8750 (99.6875) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:04:52 loss 1.0553 (1.0695) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.666, TIME@all 0.304 -epoch: [331/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0781 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0553 (1.0702) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.472, TIME@all 0.304 -epoch: [331/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:04:58 loss 1.0529 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.303 (0.304) data 0.000 (0.006) eta 0:04:52 loss 1.0804 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.841, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:04:43 loss 1.1200 (1.0605) acc 96.8750 (99.6875) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0581 (1.0710) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 842.270, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0904 (1.0599) acc 100.0000 (99.6875) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0456 (1.0727) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.322, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0498 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0429 (1.0731) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.381, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:04:43 loss 1.0475 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0560 (1.0657) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.285, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:04:43 loss 1.0534 (1.0566) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0774 (1.0699) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.504, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0786 (1.0593) acc 100.0000 (99.6875) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0613 (1.0709) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.306, TIME@all 0.304 -epoch: [332/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:04:43 loss 1.0623 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0461 (1.0674) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.702, TIME@all 0.304 -epoch: [332/350][20/50] time 0.307 (0.304) data 0.001 (0.014) eta 0:04:43 loss 1.0562 (1.0518) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.311 (0.304) data 0.000 (0.007) eta 0:04:36 loss 1.0549 (1.0635) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.451, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.012) eta 0:04:27 loss 1.0671 (1.0588) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:04:21 loss 1.0969 (1.0657) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.515, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0850 (1.0559) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0581 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.519, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:04:27 loss 1.0994 (1.0671) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:04:21 loss 1.0503 (1.0661) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.462, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0828 (1.0603) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0813 (1.0651) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.517, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:04:27 loss 1.0488 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0836 (1.0628) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.636, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0488 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0663 (1.0624) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.494, TIME@all 0.304 -epoch: [333/350][20/50] time 0.303 (0.303) data 0.000 (0.014) eta 0:04:27 loss 1.0782 (1.0639) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0776 (1.0625) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.671, TIME@all 0.304 -epoch: [333/350][20/50] time 0.302 (0.303) data 0.000 (0.013) eta 0:04:27 loss 1.0602 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:04:21 loss 1.0623 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.780, TIME@all 0.304 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0435 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0951 (1.0625) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.306, TIME@all 0.304 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0527 (1.0619) acc 100.0000 (99.6875) lr 0.000260 -epoch: [334/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0652 (1.0652) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.321, TIME@all 0.304 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:04:12 loss 1.0627 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0847 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.401, TIME@all 0.304 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0506 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0788 (1.0638) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.281, TIME@all 0.304 -epoch: [334/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0597 (1.0592) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0950 (1.0651) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.508, TIME@all 0.303 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.001 (0.014) eta 0:04:12 loss 1.0477 (1.0626) acc 100.0000 (99.8438) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0646 (1.0629) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.332, TIME@all 0.304 -epoch: [334/350][20/50] time 0.300 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0495 (1.0623) acc 100.0000 (99.8438) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.0869 (1.0658) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.488, TIME@all 0.304 -epoch: [334/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:04:12 loss 1.0486 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:04:06 loss 1.1123 (1.0607) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.698, TIME@all 0.303 -epoch: [335/350][20/50] time 0.305 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.1017 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0820 (1.0689) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.967, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0557 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0729 (1.0639) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.951, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.012) eta 0:03:57 loss 1.0678 (1.0520) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.1025 (1.0600) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.895, TIME@all 0.304 -epoch: [335/350][20/50] time 0.305 (0.305) data 0.001 (0.012) eta 0:03:57 loss 1.0545 (1.0570) acc 100.0000 (99.8438) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.0915 (1.0620) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.984, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0845 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0858 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.908, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.001 (0.013) eta 0:03:57 loss 1.0778 (1.0631) acc 100.0000 (99.8438) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:03:51 loss 1.1585 (1.0694) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.122, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.000 (0.012) eta 0:03:57 loss 1.0524 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:03:51 loss 1.0800 (1.0646) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.234, TIME@all 0.304 -epoch: [335/350][20/50] time 0.304 (0.305) data 0.000 (0.013) eta 0:03:57 loss 1.0891 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:51 loss 1.0613 (1.0650) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.086, TIME@all 0.304 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:03:41 loss 1.0726 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:03:35 loss 1.1752 (1.0650) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 844.144, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0582 (1.0536) acc 100.0000 (99.8438) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0749 (1.0646) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 844.230, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0458 (1.0510) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:03:35 loss 1.0525 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.244, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.1363 (1.0550) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0615 (1.0590) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.168, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0562 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0729 (1.0633) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.153, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:03:41 loss 1.1570 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.1174 (1.0632) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.358, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0753 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0756 (1.0640) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.308, TIME@all 0.303 -epoch: [336/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:03:41 loss 1.0620 (1.0523) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:03:35 loss 1.0577 (1.0599) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.475, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0655 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0800 (1.0697) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.842, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0521 (1.0584) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.302 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0667 (1.0662) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 845.872, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0880 (1.0686) acc 100.0000 (99.8438) lr 0.000260 -epoch: [337/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0803 (1.0733) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.909, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0700 (1.0632) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0611 (1.0663) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 845.863, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:03:26 loss 1.0619 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0526 (1.0699) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 845.880, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.001 (0.014) eta 0:03:26 loss 1.0519 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0905 (1.0727) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 846.058, TIME@all 0.303 -epoch: [337/350][20/50] time 0.303 (0.304) data 0.000 (0.014) eta 0:03:26 loss 1.0801 (1.0633) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0950 (1.0715) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 846.025, TIME@all 0.303 -epoch: [337/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:03:26 loss 1.0560 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.303 (0.303) data 0.000 (0.007) eta 0:03:20 loss 1.0774 (1.0696) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 846.078, TIME@all 0.303 -epoch: [338/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1471 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0615 (1.0668) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.566, TIME@all 0.303 -epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:03:11 loss 1.1256 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0554 (1.0601) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.622, TIME@all 0.303 -epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.0910 (1.0581) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0467 (1.0652) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.685, TIME@all 0.303 -epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1202 (1.0614) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:03:05 loss 1.0493 (1.0659) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.567, TIME@all 0.303 -epoch: [338/350][20/50] time 0.304 (0.304) data 0.001 (0.014) eta 0:03:11 loss 1.0806 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0446 (1.0595) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.799, TIME@all 0.303 -epoch: [338/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:03:11 loss 1.1176 (1.0629) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0609 (1.0682) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.600, TIME@all 0.303 -epoch: [338/350][20/50] time 0.304 (0.304) data 0.001 (0.013) eta 0:03:11 loss 1.0640 (1.0543) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0789 (1.0608) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.043, TIME@all 0.303 -epoch: [338/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:03:11 loss 1.0684 (1.0578) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:03:05 loss 1.0611 (1.0665) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.726, TIME@all 0.303 -epoch: [339/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0415 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0533 (1.0655) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.093, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0466 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0508 (1.0658) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.122, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0605 (1.0499) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0520 (1.0584) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.141, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:56 loss 1.0562 (1.0534) acc 100.0000 (99.6875) lr 0.000260 -epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:50 loss 1.0544 (1.0624) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 842.090, TIME@all 0.304 -epoch: [339/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0544 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0814 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.114, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.305) data 0.000 (0.013) eta 0:02:56 loss 1.0742 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0495 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.427, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0556 (1.0583) acc 100.0000 (99.6875) lr 0.000260 -epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0475 (1.0627) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.310, TIME@all 0.304 -epoch: [339/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:56 loss 1.0576 (1.0566) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.306 (0.304) data 0.000 (0.007) eta 0:02:50 loss 1.0454 (1.0654) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.269, TIME@all 0.304 -epoch: [340/350][20/50] time 0.306 (0.304) data 0.000 (0.011) eta 0:02:40 loss 1.0626 (1.0534) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0470 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.942, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.303) data 0.000 (0.012) eta 0:02:40 loss 1.0780 (1.0592) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.1291 (1.0661) acc 96.8750 (99.7656) lr 0.000260 -FPS@all 844.115, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0517 (1.0537) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0437 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.971, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0894 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0595 (1.0630) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.989, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0751 (1.0578) acc 100.0000 (99.6875) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0616 (1.0688) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 843.951, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.303) data 0.000 (0.013) eta 0:02:40 loss 1.0668 (1.0599) acc 100.0000 (99.6875) lr 0.000260 -epoch: [340/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:34 loss 1.0517 (1.0605) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.147, TIME@all 0.303 -epoch: [340/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0612 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.0568 (1.0570) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.397, TIME@all 0.303 -epoch: [340/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:40 loss 1.0892 (1.0586) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:34 loss 1.1269 (1.0648) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 843.980, TIME@all 0.303 -epoch: [341/350][20/50] time 0.305 (0.305) data 0.000 (0.012) eta 0:02:26 loss 1.0789 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:20 loss 1.1212 (1.0646) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.057, TIME@all 0.304 -epoch: [341/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0598 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:20 loss 1.0537 (1.0663) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.092, TIME@all 0.304 -epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.1190 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0751 (1.0633) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.126, TIME@all 0.304 -epoch: [341/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.1517 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0679 (1.0708) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.080, TIME@all 0.304 -epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0520 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:20 loss 1.0891 (1.0665) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.077, TIME@all 0.304 -epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.014) eta 0:02:26 loss 1.0631 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0564 (1.0632) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.274, TIME@all 0.304 -epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0917 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.0598 (1.0673) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 843.226, TIME@all 0.304 -epoch: [341/350][20/50] time 0.305 (0.304) data 0.000 (0.013) eta 0:02:26 loss 1.0829 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:02:19 loss 1.1178 (1.0708) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.474, TIME@all 0.304 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0532 (1.0567) acc 100.0000 (99.8438) lr 0.000260 -epoch: [342/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0885 (1.0630) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.753, TIME@all 0.303 -epoch: [342/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0464 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0717 (1.0656) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.789, TIME@all 0.303 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:02:10 loss 1.0506 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.302 (0.304) data 0.001 (0.006) eta 0:02:04 loss 1.0597 (1.0651) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.829, TIME@all 0.303 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.012) eta 0:02:10 loss 1.0534 (1.0542) acc 100.0000 (99.8438) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.001 (0.006) eta 0:02:04 loss 1.1029 (1.0685) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.744, TIME@all 0.303 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0877 (1.0562) acc 100.0000 (99.8438) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.0855 (1.0670) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 843.752, TIME@all 0.303 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.001 (0.013) eta 0:02:10 loss 1.0494 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:02:04 loss 1.0859 (1.0630) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.920, TIME@all 0.303 -epoch: [342/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:02:10 loss 1.0537 (1.0610) acc 100.0000 (99.8438) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:02:04 loss 1.1037 (1.0760) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 844.080, TIME@all 0.303 -epoch: [342/350][20/50] time 0.305 (0.304) data 0.001 (0.013) eta 0:02:10 loss 1.0542 (1.0534) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:02:04 loss 1.0975 (1.0747) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 843.941, TIME@all 0.303 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.2145 (1.0600) acc 96.8750 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0720 (1.0634) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.588, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.1550 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0514 (1.0673) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.503, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.012) eta 0:01:55 loss 1.1553 (1.0548) acc 96.8750 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.301 (0.304) data 0.000 (0.006) eta 0:01:49 loss 1.0487 (1.0656) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 842.642, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1276 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0684 (1.0682) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.527, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1111 (1.0588) acc 100.0000 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0706 (1.0639) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.672, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.0823 (1.0590) acc 100.0000 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0598 (1.0621) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.525, TIME@all 0.304 -epoch: [343/350][20/50] time 0.301 (0.304) data 0.000 (0.014) eta 0:01:55 loss 1.1264 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0555 (1.0626) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.725, TIME@all 0.304 -epoch: [343/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:55 loss 1.1919 (1.0589) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:01:49 loss 1.0586 (1.0660) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.824, TIME@all 0.304 -epoch: [344/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0506 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0712 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.653, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:40 loss 1.0637 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:01:34 loss 1.0491 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.554, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0525 (1.0501) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0953 (1.0599) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.615, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0882 (1.0511) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0507 (1.0632) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.582, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0846 (1.0504) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0737 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.562, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0594 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0511 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.767, TIME@all 0.303 -epoch: [344/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:01:40 loss 1.0529 (1.0548) acc 100.0000 (99.8438) lr 0.000260 -epoch: [344/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0440 (1.0636) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.966, TIME@all 0.303 -epoch: [344/350][20/50] time 0.302 (0.304) data 0.000 (0.014) eta 0:01:40 loss 1.0523 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:01:34 loss 1.0492 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.719, TIME@all 0.303 -epoch: [345/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.0863 (1.0556) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:18 loss 1.0439 (1.0625) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.171, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.0995 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:18 loss 1.0462 (1.0604) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.121, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:01:25 loss 1.1178 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:18 loss 1.0489 (1.0650) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.194, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0824 (1.0504) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.001 (0.007) eta 0:01:18 loss 1.1278 (1.0634) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 844.137, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0948 (1.0522) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.303) data 0.000 (0.007) eta 0:01:18 loss 1.0696 (1.0621) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.336, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0892 (1.0630) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0532 (1.0677) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.143, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.0968 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0832 (1.0620) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 844.284, TIME@all 0.303 -epoch: [345/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:01:25 loss 1.2745 (1.0650) acc 96.8750 (99.8438) lr 0.000260 -epoch: [345/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:18 loss 1.0714 (1.0690) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 844.451, TIME@all 0.303 -epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0566 (1.0504) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1366 (1.0618) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.723, TIME@all 0.303 -epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0639 (1.0583) acc 100.0000 (99.8438) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1625 (1.0638) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 843.796, TIME@all 0.303 -epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0552 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:03 loss 1.0716 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.717, TIME@all 0.303 -epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:09 loss 1.0594 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.0958 (1.0577) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 843.793, TIME@all 0.303 -epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.013) eta 0:01:09 loss 1.0578 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.007) eta 0:01:03 loss 1.1800 (1.0616) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 843.944, TIME@all 0.303 -epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0536 (1.0549) acc 100.0000 (99.8438) lr 0.000260 -epoch: [346/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.0727 (1.0590) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 844.088, TIME@all 0.303 -epoch: [346/350][20/50] time 0.302 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0531 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.000 (0.006) eta 0:01:03 loss 1.1451 (1.0627) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.906, TIME@all 0.303 -epoch: [346/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:01:09 loss 1.0551 (1.0513) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.304 (0.304) data 0.001 (0.006) eta 0:01:03 loss 1.0725 (1.0603) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.731, TIME@all 0.303 -epoch: [347/350][20/50] time 0.305 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0459 (1.0593) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0824 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.361, TIME@all 0.304 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0529 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.308 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0593 (1.0683) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.337, TIME@all 0.304 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.012) eta 0:00:54 loss 1.0488 (1.0536) acc 100.0000 (99.8438) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.006) eta 0:00:48 loss 1.0723 (1.0645) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.417, TIME@all 0.304 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0476 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0791 (1.0668) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.326, TIME@all 0.304 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0474 (1.0509) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0682 (1.0637) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.347, TIME@all 0.304 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.014) eta 0:00:54 loss 1.0510 (1.0512) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.308 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0831 (1.0639) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.526, TIME@all 0.303 -epoch: [347/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0541 (1.0540) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.001 (0.007) eta 0:00:48 loss 1.0650 (1.0689) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.638, TIME@all 0.303 -epoch: [347/350][20/50] time 0.304 (0.304) data 0.000 (0.013) eta 0:00:54 loss 1.0512 (1.0600) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.307 (0.304) data 0.000 (0.007) eta 0:00:48 loss 1.0734 (1.0767) acc 100.0000 (99.5312) lr 0.000260 -FPS@all 843.505, TIME@all 0.303 -epoch: [348/350][20/50] time 0.306 (0.304) data 0.000 (0.012) eta 0:00:39 loss 1.2009 (1.0585) acc 96.8750 (99.8438) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:00:33 loss 1.0688 (1.0711) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 842.848, TIME@all 0.304 -epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.012) eta 0:00:39 loss 1.0993 (1.0502) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.006) eta 0:00:33 loss 1.0607 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.955, TIME@all 0.304 -epoch: [348/350][20/50] time 0.306 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1287 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0657 (1.0636) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.883, TIME@all 0.304 -epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1586 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0507 (1.0664) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.887, TIME@all 0.304 -epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1240 (1.0571) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0503 (1.0677) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.878, TIME@all 0.304 -epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1111 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0784 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.029, TIME@all 0.304 -epoch: [348/350][20/50] time 0.307 (0.304) data 0.000 (0.014) eta 0:00:39 loss 1.0875 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.305 (0.304) data 0.001 (0.007) eta 0:00:33 loss 1.0564 (1.0620) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.101, TIME@all 0.304 -epoch: [348/350][20/50] time 0.308 (0.304) data 0.000 (0.013) eta 0:00:39 loss 1.1682 (1.0571) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.303 (0.304) data 0.000 (0.007) eta 0:00:33 loss 1.0693 (1.0653) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.205, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0618 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0585 (1.0622) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.800, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0487 (1.0562) acc 100.0000 (99.8438) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0740 (1.0619) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.917, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0497 (1.0596) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0446 (1.0663) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.756, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.012) eta 0:00:24 loss 1.0556 (1.0591) acc 100.0000 (99.8438) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:00:18 loss 1.0548 (1.0634) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 842.739, TIME@all 0.304 -epoch: [349/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0463 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.006) eta 0:00:18 loss 1.0492 (1.0653) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 842.818, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0612 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.1281 (1.0657) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 842.767, TIME@all 0.304 -epoch: [349/350][20/50] time 0.307 (0.305) data 0.000 (0.014) eta 0:00:24 loss 1.0551 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.300 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0709 (1.0667) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 842.956, TIME@all 0.304 -epoch: [349/350][20/50] time 0.308 (0.305) data 0.000 (0.013) eta 0:00:24 loss 1.0540 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.301 (0.304) data 0.000 (0.007) eta 0:00:18 loss 1.0612 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.150, TIME@all 0.304 -epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.012) eta 0:00:09 loss 1.0615 (1.0651) acc 100.0000 (99.6875) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0467 (1.0691) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 843.075, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.302 (0.304) data 0.001 (0.012) eta 0:00:09 loss 1.0810 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0471 (1.0675) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.058, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0634 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.006) eta 0:00:03 loss 1.0873 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 843.146, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.1015 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.001 (0.007) eta 0:00:03 loss 1.0493 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.089, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0629 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0624 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.078, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.301 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.1251 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0603 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 843.389, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.303 (0.304) data 0.000 (0.013) eta 0:00:09 loss 1.0980 (1.0542) acc 100.0000 (99.8438) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0488 (1.0639) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.228, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.303 (0.304) data 0.001 (0.013) eta 0:00:09 loss 1.0610 (1.0521) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.302 (0.304) data 0.000 (0.007) eta 0:00:03 loss 1.0705 (1.0638) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 843.298, TIME@all 0.304 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 15913-by-512 matrix -Speed: 0.0301 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.3% -CMC curve -Rank-1 : 93.1% -Rank-5 : 97.6% -Rank-10 : 98.3% -Rank-20 : 99.0% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:44:48 -FPS@all 842.343, TIME@all 0.304 -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0302 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.2% -CMC curve -Rank-1 : 93.0% -Rank-5 : 97.5% -Rank-10 : 98.5% -Rank-20 : 99.0% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:44:53 -FPS@all 842.416, TIME@all 0.304 -Done, obtained 15913-by-512 matrix -Speed: 0.0301 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.3% -CMC curve -Rank-1 : 93.1% -Rank-5 : 97.6% -Rank-10 : 98.4% -Rank-20 : 99.2% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:44:55 -FPS@all 842.380, TIME@all 0.304 -Done, obtained 15913-by-512 matrix -Speed: 0.0302 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.2% -CMC curve -Rank-1 : 93.0% -Rank-5 : 97.4% -Rank-10 : 98.4% -Rank-20 : 99.1% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:44:57 -FPS@all 842.697, TIME@all 0.304 -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0312 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.2% -CMC curve -Rank-1 : 92.8% -Rank-5 : 97.5% -Rank-10 : 98.5% -Rank-20 : 99.0% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:45:00 -FPS@all 842.357, TIME@all 0.304 -Done, obtained 15913-by-512 matrix -Speed: 0.0310 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.2% -CMC curve -Rank-1 : 92.7% -Rank-5 : 97.6% -Rank-10 : 98.5% -Rank-20 : 99.1% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:45:00 -FPS@all 842.363, TIME@all 0.304 -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0319 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.2% -CMC curve -Rank-1 : 92.8% -Rank-5 : 97.5% -Rank-10 : 98.5% -Rank-20 : 99.1% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:45:01 -FPS@all 842.505, TIME@all 0.304 -Done, obtained 15913-by-512 matrix -Speed: 0.0333 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 80.3% -CMC curve -Rank-1 : 93.0% -Rank-5 : 97.5% -Rank-10 : 98.4% -Rank-20 : 99.1% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:45:04 -FPS@all 842.553, TIME@all 0.304 -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -THPModule_npu_shutdown success. -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 7 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 6 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 0 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 2 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 3 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 1 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 4 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Show configuration -adam: - beta1: 0.9 - beta2: 0.999 -addr: 127.0.0.1 -amp: True -cuhk03: - classic_split: False - labeled_images: False - use_metric_cuhk03: False -data: - combineall: False - height: 256 - k_tfm: 1 - load_train_targets: False - norm_mean: [0.485, 0.456, 0.406] - norm_std: [0.229, 0.224, 0.225] - root: reid-data - save_dir: log/osnet_x1_0_market1501_softmax - sources: ['market1501'] - split_id: 0 - targets: ['market1501'] - transforms: ['random_flip', 'random_crop', 'random_patch'] - type: image - width: 128 - workers: 4 -device_num: 8 -ignore_classifer: False -local_rank: 5 -loss: - name: softmax - softmax: - label_smooth: True - triplet: - margin: 0.3 - weight_t: 1.0 - weight_x: 0.0 -market1501: - use_500k_distractors: False -model: - load_weights: - name: osnet_x1_0 - pretrained: False - resume: -rmsprop: - alpha: 0.99 -sampler: - num_cams: 1 - num_datasets: 1 - num_instances: 4 - train_sampler: RandomSampler - train_sampler_t: RandomSampler -sgd: - dampening: 0.0 - momentum: 0.9 - nesterov: False -test: - batch_size: 300 - dist_metric: euclidean - eval_freq: -1 - evaluate: False - normalize_feature: False - ranks: [1, 5, 10, 20] - rerank: False - start_eval: 300 - visrank: False - visrank_topk: 10 -train: - base_lr_mult: 0.1 - batch_size: 32 - fixbase_epoch: 0 - gamma: 0.1 - lr: 0.26 - lr_scheduler: multi_step - max_epoch: 350 - new_layers: ['classifier'] - open_layers: ['classifier'] - optim: sgd - print_freq: 20 - seed: 1 - staged_lr: False - start_epoch: 0 - stepsize: [150, 225, 300] - weight_decay: 0.0005 -use_gpu: False -use_npu: True -video: - pooling_method: avg - sample_method: evenly - seq_len: 15 - -Collecting env info ... -** System info ** -PyTorch version: 1.8.1+ascend.rc2 -Is debug build: False -CUDA used to build PyTorch: None -ROCM used to build PyTorch: N/A - -OS: CentOS Linux 7 (AltArch) (aarch64) -GCC version: (GCC) 7.3.0 -Clang version: 3.9.1 (tags/RELEASE_391/final) -CMake version: version 3.18.6 - -Python version: 3.7 (64-bit runtime) -Is CUDA available: False -CUDA runtime version: No CUDA -GPU models and configuration: No CUDA -Nvidia driver version: No CUDA -cuDNN version: No CUDA -HIP runtime version: N/A -MIOpen runtime version: N/A - -Versions of relevant libraries: -[pip3] numpy==1.20.0 -[pip3] torch==1.8.1+ascend.rc2.20220505 -[pip3] torch-npu==1.8.1rc2.post20220505 -[pip3] torchreid==1.4.0 -[conda] numpy 1.20.0 -[conda] torch 1.8.1+ascend.rc2.20220505 -[conda] torch-npu 1.8.1rc2.post20220505 -[conda] torchreid 1.4.0 - Pillow (8.4.0) - -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Selected optimization level O2: FP16 training with FP32 batchnorm and FP32 master weights. - -Defaults for this optimization level are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : dynamic -combine_grad : None -combine_ddp : None -ddp_replica_count : 4 -check_combined_tensors : None -user_cast_preferred : None -Processing user overrides (additional kwargs that are not None)... -After processing overrides, optimization options are: -enabled : True -opt_level : O2 -cast_model_type : torch.float16 -patch_torch_functions : False -keep_batchnorm_fp32 : True -master_weights : True -loss_scale : dynamic -combine_grad : True -combine_ddp : None -ddp_replica_count : 4 -check_combined_tensors : None -user_cast_preferred : None -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -Building train transforms ... -+ resize to 256x128 -+ random flip -+ random crop (enlarge to 288x144 and crop 256x128) -+ random patch -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -Building test transforms ... -+ resize to 256x128 -+ to torch tensor of range [0, 1] -+ normalization (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) -=> Loading train (source) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- -=> Loading test (target) dataset -=> Loaded Market1501 - ---------------------------------------- - subset | # ids | # images | # cameras - ---------------------------------------- - train | 751 | 12936 | 6 - query | 750 | 3368 | 6 - gallery | 751 | 15913 | 6 - ---------------------------------------- - - - **************** Summary **************** - source : ['market1501'] - # source datasets : 1 - # source ids : 751 - # source images : 12936 - # source cameras : 6 - target : ['market1501'] - ***************************************** - - -Building model: osnet_x1_0 -Model complexity: params=2,193,616 flops=978,878,352 -Use npu fused optimizer -Building softmax-engine for image-reid -=> Start training -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:1063] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters. This flag results in an extra traversal of the autograd graph every iteration, which can adversely affect performance. If your model indeed never has any unused parameters, consider turning this flag off. Note that this warning may be a false positive your model has flow control causing later iterations to have unused parameters. (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -[W reducer.cpp:399] Warning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance. -grad.sizes() = [751, 512], strides() = [512, 1] -bucket_view.sizes() = [385024], strides() = [1] (function operator()) -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.316 (0.311) data 0.000 (0.020) eta 1:30:42 loss 6.6282 (6.6135) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/350][40/50] time 0.315 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.0348 (6.4896) acc 6.2500 (0.8594) lr 0.260000 -FPS@all 821.678, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.8040 (6.5953) acc 0.0000 (0.4688) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1929 (6.4627) acc 0.0000 (0.9375) lr 0.260000 -FPS@all 821.595, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:43 loss 6.6438 (6.6055) acc 0.0000 (0.3125) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1450 (6.4875) acc 6.2500 (0.5469) lr 0.260000 -FPS@all 821.637, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.019) eta 1:30:43 loss 6.2564 (6.6608) acc 3.1250 (0.6250) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.2591 (6.5006) acc 3.1250 (0.9375) lr 0.260000 -FPS@all 821.562, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.4853 (6.6085) acc 0.0000 (0.4688) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.1861 (6.4820) acc 0.0000 (0.8594) lr 0.260000 -FPS@all 821.564, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.7360 (6.6453) acc 0.0000 (0.6250) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:43 loss 6.1298 (6.4847) acc 0.0000 (0.9375) lr 0.260000 -FPS@all 821.594, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:44 loss 6.5267 (6.6343) acc 0.0000 (0.7812) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.001 (0.010) eta 1:30:44 loss 5.9071 (6.4698) acc 6.2500 (1.0156) lr 0.260000 -FPS@all 821.580, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 32768.0 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -group num: 1 -epoch: [1/350][20/50] time 0.317 (0.311) data 0.000 (0.020) eta 1:30:43 loss 6.4037 (6.6935) acc 0.0000 (0.4688) lr 0.260000 -epoch: [1/350][40/50] time 0.314 (0.312) data 0.000 (0.010) eta 1:30:44 loss 6.2472 (6.5596) acc 0.0000 (0.7031) lr 0.260000 -FPS@all 821.601, TIME@all 0.312 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 6.0594 (5.7936) acc 3.1250 (3.2812) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.7203 (5.7094) acc 6.2500 (3.9844) lr 0.260000 -FPS@all 819.110, TIME@all 0.313 -epoch: [2/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 1:31:09 loss 5.9182 (5.6758) acc 0.0000 (3.7500) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:53 loss 5.6438 (5.6485) acc 9.3750 (4.6875) lr 0.260000 -FPS@all 819.211, TIME@all 0.312 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.5824 (5.7873) acc 12.5000 (3.1250) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.3965 (5.6825) acc 3.1250 (4.3750) lr 0.260000 -FPS@all 819.078, TIME@all 0.313 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.5807 (5.6932) acc 0.0000 (4.3750) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 5.5829 (5.6599) acc 6.2500 (3.6719) lr 0.260000 -FPS@all 819.054, TIME@all 0.313 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:31:10 loss 6.0790 (5.7108) acc 0.0000 (4.0625) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:54 loss 5.3312 (5.6364) acc 6.2500 (4.2969) lr 0.260000 -FPS@all 819.147, TIME@all 0.313 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:31:10 loss 5.8013 (5.6971) acc 3.1250 (4.6875) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:30:54 loss 5.3562 (5.6617) acc 9.3750 (4.5312) lr 0.260000 -FPS@all 819.131, TIME@all 0.313 -epoch: [2/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 1:31:12 loss 5.9211 (5.7517) acc 3.1250 (4.0625) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:30:54 loss 5.2705 (5.6378) acc 9.3750 (4.8438) lr 0.260000 -FPS@all 819.092, TIME@all 0.313 -epoch: [2/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:31:10 loss 5.7639 (5.7057) acc 3.1250 (4.3750) lr 0.260000 -epoch: [2/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:30:54 loss 6.0995 (5.6769) acc 3.1250 (4.6875) lr 0.260000 -FPS@all 819.107, TIME@all 0.313 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.2832 (5.0634) acc 12.5000 (9.8438) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.9239 (5.0338) acc 12.5000 (10.3906) lr 0.260000 -FPS@all 821.691, TIME@all 0.312 -epoch: [3/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:30:26 loss 5.0421 (5.0929) acc 9.3750 (8.1250) lr 0.260000 -epoch: [3/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:30:18 loss 5.1250 (5.1040) acc 6.2500 (8.8281) lr 0.260000 -FPS@all 821.794, TIME@all 0.312 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3481 (5.0394) acc 9.3750 (8.5938) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 5.3156 (5.0389) acc 6.2500 (9.6875) lr 0.260000 -FPS@all 821.637, TIME@all 0.312 -epoch: [3/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:30:28 loss 5.3440 (5.0542) acc 6.2500 (7.8125) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 5.1863 (5.0536) acc 6.2500 (8.5938) lr 0.260000 -FPS@all 821.704, TIME@all 0.312 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3779 (5.0746) acc 3.1250 (8.5938) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.5517 (5.0395) acc 18.7500 (10.0000) lr 0.260000 -FPS@all 821.575, TIME@all 0.312 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.3214 (5.0822) acc 3.1250 (7.1875) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.8877 (5.0807) acc 15.6250 (7.5000) lr 0.260000 -FPS@all 821.692, TIME@all 0.312 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:30:28 loss 5.1941 (5.0012) acc 6.2500 (6.5625) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:30:19 loss 4.5551 (4.9956) acc 12.5000 (8.9844) lr 0.260000 -FPS@all 821.684, TIME@all 0.312 -epoch: [3/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:30:28 loss 5.2215 (5.0933) acc 9.3750 (9.5312) lr 0.260000 -epoch: [3/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:30:19 loss 4.7463 (5.0586) acc 12.5000 (9.9219) lr 0.260000 -FPS@all 821.633, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:30:08 loss 4.5543 (4.3096) acc 18.7500 (19.6875) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:03 loss 4.8111 (4.4200) acc 15.6250 (18.2812) lr 0.260000 -FPS@all 821.298, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:30:08 loss 4.6653 (4.3899) acc 15.6250 (17.8125) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.4343 (4.4955) acc 21.8750 (17.6562) lr 0.260000 -FPS@all 821.246, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.2725 (4.3375) acc 25.0000 (18.7500) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.7870 (4.4286) acc 18.7500 (17.8125) lr 0.260000 -FPS@all 821.185, TIME@all 0.312 -epoch: [4/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5120 (4.3763) acc 9.3750 (17.1875) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.5298 (4.4602) acc 18.7500 (18.0469) lr 0.260000 -FPS@all 821.231, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.6118 (4.3939) acc 15.6250 (18.1250) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:30:04 loss 4.8010 (4.4485) acc 18.7500 (17.9688) lr 0.260000 -FPS@all 821.188, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5778 (4.4145) acc 21.8750 (17.9688) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:30:04 loss 4.2522 (4.4577) acc 25.0000 (17.4219) lr 0.260000 -FPS@all 821.161, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.1779 (4.4533) acc 21.8750 (17.8125) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.7549 (4.4869) acc 9.3750 (17.1094) lr 0.260000 -FPS@all 821.213, TIME@all 0.312 -epoch: [4/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:30:09 loss 4.5193 (4.3964) acc 9.3750 (15.7812) lr 0.260000 -epoch: [4/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:30:04 loss 4.0433 (4.4299) acc 31.2500 (17.6562) lr 0.260000 -FPS@all 821.152, TIME@all 0.312 -epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:53 loss 3.8829 (3.7926) acc 31.2500 (30.0000) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.0074 (3.8962) acc 25.0000 (29.3750) lr 0.260000 -FPS@all 822.299, TIME@all 0.311 -epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:52 loss 3.4718 (3.8024) acc 43.7500 (30.0000) lr 0.260000 -epoch: [5/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:29:40 loss 3.6481 (3.8641) acc 37.5000 (28.9844) lr 0.260000 -FPS@all 822.400, TIME@all 0.311 -epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:29:55 loss 4.1549 (3.7249) acc 18.7500 (31.0938) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:29:41 loss 4.3441 (3.8644) acc 15.6250 (28.3594) lr 0.260000 -FPS@all 822.233, TIME@all 0.311 -epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:54 loss 3.8879 (3.7821) acc 25.0000 (30.1562) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.6259 (3.8353) acc 43.7500 (31.2500) lr 0.260000 -FPS@all 822.243, TIME@all 0.311 -epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:54 loss 4.4040 (3.7438) acc 12.5000 (26.8750) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.7219 (3.8507) acc 28.1250 (28.0469) lr 0.260000 -FPS@all 822.306, TIME@all 0.311 -epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:53 loss 4.3460 (3.8087) acc 25.0000 (27.8125) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.2156 (3.8984) acc 37.5000 (28.7500) lr 0.260000 -FPS@all 822.309, TIME@all 0.311 -epoch: [5/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:55 loss 3.5324 (3.8007) acc 37.5000 (30.9375) lr 0.260000 -epoch: [5/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:29:41 loss 3.9986 (3.8975) acc 18.7500 (29.2188) lr 0.260000 -FPS@all 822.305, TIME@all 0.311 -epoch: [5/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:29:54 loss 4.3092 (3.8331) acc 15.6250 (29.8438) lr 0.260000 -epoch: [5/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:29:41 loss 4.0275 (3.8858) acc 31.2500 (29.7656) lr 0.260000 -FPS@all 822.267, TIME@all 0.311 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:29:30 loss 3.9603 (3.4243) acc 28.1250 (40.7812) lr 0.260000 -epoch: [6/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:29:31 loss 3.7473 (3.4712) acc 34.3750 (40.0000) lr 0.260000 -FPS@all 821.414, TIME@all 0.312 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:29:29 loss 3.9790 (3.3634) acc 18.7500 (38.4375) lr 0.260000 -epoch: [6/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:29:30 loss 3.8342 (3.4920) acc 25.0000 (36.9531) lr 0.260000 -FPS@all 821.463, TIME@all 0.312 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7156 (3.2911) acc 34.3750 (40.9375) lr 0.260000 -epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.3102 (3.4258) acc 40.6250 (39.6875) lr 0.260000 -FPS@all 821.298, TIME@all 0.312 -epoch: [6/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:29:30 loss 3.5310 (3.3215) acc 46.8750 (41.8750) lr 0.260000 -epoch: [6/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:29:31 loss 3.5656 (3.4785) acc 34.3750 (37.8906) lr 0.260000 -FPS@all 821.316, TIME@all 0.312 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7731 (3.3322) acc 40.6250 (40.3125) lr 0.260000 -epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.5444 (3.4172) acc 40.6250 (40.1562) lr 0.260000 -FPS@all 821.248, TIME@all 0.312 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.7150 (3.3124) acc 31.2500 (42.6562) lr 0.260000 -epoch: [6/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.4702 (3.4333) acc 34.3750 (38.9062) lr 0.260000 -FPS@all 821.333, TIME@all 0.312 -epoch: [6/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:29:30 loss 3.6914 (3.3798) acc 34.3750 (38.7500) lr 0.260000 -epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:29:31 loss 3.8998 (3.5084) acc 34.3750 (36.0938) lr 0.260000 -FPS@all 821.341, TIME@all 0.312 -epoch: [6/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:29:30 loss 3.4229 (3.2996) acc 43.7500 (43.1250) lr 0.260000 -epoch: [6/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:29:31 loss 3.7600 (3.4180) acc 40.6250 (40.2344) lr 0.260000 -FPS@all 821.329, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.014) eta 1:29:40 loss 2.6181 (2.8738) acc 59.3750 (55.6250) lr 0.260000 -epoch: [7/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:29:24 loss 3.1618 (3.0069) acc 50.0000 (51.4844) lr 0.260000 -FPS@all 820.658, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.014) eta 1:29:40 loss 2.6422 (2.8396) acc 59.3750 (52.6562) lr 0.260000 -epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.2460 (2.9835) acc 50.0000 (50.7031) lr 0.260000 -FPS@all 820.598, TIME@all 0.312 -epoch: [7/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:29:41 loss 2.9682 (2.9077) acc 56.2500 (53.5938) lr 0.260000 -epoch: [7/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.8760 (3.0461) acc 28.1250 (50.5469) lr 0.260000 -FPS@all 820.447, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:29:40 loss 3.0837 (2.8961) acc 50.0000 (53.1250) lr 0.260000 -epoch: [7/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.3221 (3.0226) acc 43.7500 (51.1719) lr 0.260000 -FPS@all 820.540, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 1:29:41 loss 3.0217 (2.8466) acc 50.0000 (53.4375) lr 0.260000 -epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:29:25 loss 3.4676 (2.9847) acc 43.7500 (51.1719) lr 0.260000 -FPS@all 820.478, TIME@all 0.312 -epoch: [7/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:29:41 loss 3.2219 (2.8566) acc 43.7500 (53.5938) lr 0.260000 -epoch: [7/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.3742 (2.9948) acc 53.1250 (52.0312) lr 0.260000 -FPS@all 820.515, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:29:40 loss 3.0862 (2.8851) acc 28.1250 (50.0000) lr 0.260000 -epoch: [7/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:29:25 loss 3.4040 (3.0138) acc 46.8750 (48.9062) lr 0.260000 -FPS@all 820.538, TIME@all 0.312 -epoch: [7/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 1:29:41 loss 2.7115 (2.8474) acc 43.7500 (52.8125) lr 0.260000 -epoch: [7/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:29:25 loss 3.2168 (3.0193) acc 40.6250 (49.5312) lr 0.260000 -FPS@all 820.532, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:58 loss 2.8560 (2.6150) acc 59.3750 (63.1250) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.8085 (2.7530) acc 56.2500 (58.2812) lr 0.260000 -FPS@all 821.782, TIME@all 0.312 -epoch: [8/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:29:00 loss 3.1258 (2.6251) acc 50.0000 (59.5312) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.9904 (2.7642) acc 50.0000 (56.2500) lr 0.260000 -FPS@all 821.743, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:28:59 loss 3.2261 (2.5946) acc 43.7500 (61.8750) lr 0.260000 -epoch: [8/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:28:59 loss 2.8836 (2.7469) acc 50.0000 (58.2031) lr 0.260000 -FPS@all 821.580, TIME@all 0.312 -epoch: [8/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:28:59 loss 3.1327 (2.5782) acc 53.1250 (62.0312) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:59 loss 2.7518 (2.7192) acc 59.3750 (58.6719) lr 0.260000 -FPS@all 821.669, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:28:59 loss 3.1808 (2.6771) acc 43.7500 (57.6562) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:28:58 loss 2.8134 (2.8056) acc 50.0000 (55.0781) lr 0.260000 -FPS@all 821.619, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:28:57 loss 3.0446 (2.5873) acc 53.1250 (59.5312) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:28:57 loss 3.0669 (2.7482) acc 46.8750 (56.3281) lr 0.260000 -FPS@all 821.778, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:59 loss 2.9624 (2.5745) acc 40.6250 (60.9375) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.9526 (2.7848) acc 59.3750 (56.4844) lr 0.260000 -FPS@all 821.656, TIME@all 0.312 -epoch: [8/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:59 loss 3.0653 (2.6125) acc 65.6250 (61.2500) lr 0.260000 -epoch: [8/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:28:58 loss 2.6907 (2.7866) acc 59.3750 (55.9375) lr 0.260000 -FPS@all 821.671, TIME@all 0.312 -epoch: [9/350][20/50] time 0.313 (0.311) data 0.000 (0.013) eta 1:28:33 loss 3.2726 (2.4585) acc 53.1250 (66.2500) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.0010 (2.6279) acc 50.0000 (62.9688) lr 0.260000 -FPS@all 822.341, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.2206 (2.3918) acc 53.1250 (68.4375) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.1333 (2.5713) acc 50.0000 (63.9062) lr 0.260000 -FPS@all 822.369, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.0911 (2.4511) acc 46.8750 (65.6250) lr 0.260000 -epoch: [9/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8350 (2.5984) acc 56.2500 (62.3438) lr 0.260000 -FPS@all 822.282, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 2.9704 (2.4047) acc 53.1250 (67.8125) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8338 (2.5180) acc 56.2500 (64.1406) lr 0.260000 -FPS@all 822.243, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.1379 (2.4459) acc 53.1250 (66.8750) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.8686 (2.5407) acc 59.3750 (63.8281) lr 0.260000 -FPS@all 822.252, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 2.8515 (2.4064) acc 65.6250 (68.5938) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.9643 (2.5307) acc 53.1250 (64.2188) lr 0.260000 -FPS@all 822.313, TIME@all 0.311 -epoch: [9/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.0585 (2.5614) acc 37.5000 (60.4688) lr 0.260000 -epoch: [9/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:28:34 loss 2.7029 (2.6364) acc 65.6250 (60.2344) lr 0.260000 -FPS@all 822.290, TIME@all 0.311 -epoch: [9/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:28:31 loss 3.3398 (2.4733) acc 46.8750 (65.0000) lr 0.260000 -epoch: [9/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:28:34 loss 3.2641 (2.5778) acc 34.3750 (63.1250) lr 0.260000 -FPS@all 822.314, TIME@all 0.311 -epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:43 loss 2.3517 (2.2324) acc 68.7500 (70.6250) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:28:31 loss 2.4602 (2.3565) acc 75.0000 (68.9062) lr 0.260000 -FPS@all 821.370, TIME@all 0.312 -epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.6665 (2.2341) acc 71.8750 (71.4062) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4884 (2.3236) acc 75.0000 (69.7656) lr 0.260000 -FPS@all 821.271, TIME@all 0.312 -epoch: [10/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.5749 (2.1301) acc 62.5000 (74.6875) lr 0.260000 -epoch: [10/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:28:33 loss 2.2472 (2.2847) acc 78.1250 (71.6406) lr 0.260000 -FPS@all 821.225, TIME@all 0.312 -epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:28:43 loss 2.4498 (2.1611) acc 71.8750 (73.2812) lr 0.260000 -epoch: [10/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:28:32 loss 2.0609 (2.2643) acc 78.1250 (71.2500) lr 0.260000 -FPS@all 821.249, TIME@all 0.312 -epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.6028 (2.2636) acc 59.3750 (70.0000) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4973 (2.3759) acc 71.8750 (68.2812) lr 0.260000 -FPS@all 821.222, TIME@all 0.312 -epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.5421 (2.1689) acc 65.6250 (71.4062) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.3444 (2.3500) acc 65.6250 (66.3281) lr 0.260000 -FPS@all 821.268, TIME@all 0.312 -epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.3957 (2.1793) acc 59.3750 (72.1875) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.4514 (2.3398) acc 78.1250 (68.9844) lr 0.260000 -FPS@all 821.277, TIME@all 0.312 -epoch: [10/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:43 loss 2.2044 (2.1960) acc 68.7500 (72.8125) lr 0.260000 -epoch: [10/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:28:32 loss 2.5738 (2.3373) acc 59.3750 (68.5938) lr 0.260000 -FPS@all 821.241, TIME@all 0.312 -epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:26 loss 2.3666 (2.0413) acc 68.7500 (77.6562) lr 0.260000 -epoch: [11/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0087 (2.1680) acc 75.0000 (75.0000) lr 0.260000 -FPS@all 821.105, TIME@all 0.312 -epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:26 loss 2.1604 (2.0000) acc 71.8750 (80.0000) lr 0.260000 -epoch: [11/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0525 (2.1788) acc 84.3750 (75.0781) lr 0.260000 -FPS@all 821.184, TIME@all 0.312 -epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.011) eta 1:28:27 loss 2.4809 (2.0056) acc 78.1250 (80.0000) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.0819 (2.1647) acc 75.0000 (75.0781) lr 0.260000 -FPS@all 821.026, TIME@all 0.312 -epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:28 loss 1.8513 (1.9640) acc 84.3750 (80.9375) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:18 loss 2.3437 (2.1515) acc 71.8750 (77.0312) lr 0.260000 -FPS@all 820.962, TIME@all 0.312 -epoch: [11/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.5034 (1.9836) acc 62.5000 (80.6250) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2583 (2.1649) acc 75.0000 (75.3125) lr 0.260000 -FPS@all 821.041, TIME@all 0.312 -epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.4249 (2.0324) acc 65.6250 (77.3438) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.1741 (2.1452) acc 71.8750 (75.2344) lr 0.260000 -FPS@all 821.095, TIME@all 0.312 -epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 1.9815 (1.9537) acc 78.1250 (80.4688) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2618 (2.1193) acc 71.8750 (76.0156) lr 0.260000 -FPS@all 821.061, TIME@all 0.312 -epoch: [11/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:27 loss 2.2706 (2.0041) acc 65.6250 (78.4375) lr 0.260000 -epoch: [11/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:28:17 loss 2.2561 (2.1261) acc 75.0000 (76.2500) lr 0.260000 -FPS@all 821.005, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:11 loss 2.3362 (1.9903) acc 71.8750 (79.6875) lr 0.260000 -epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:28:00 loss 2.3193 (2.1579) acc 59.3750 (75.1562) lr 0.260000 -FPS@all 821.060, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:10 loss 2.6404 (1.9881) acc 71.8750 (79.2188) lr 0.260000 -epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:28:00 loss 2.2368 (2.1016) acc 75.0000 (76.1719) lr 0.260000 -FPS@all 821.081, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:10 loss 2.1621 (1.8695) acc 81.2500 (83.2812) lr 0.260000 -epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.3304 (2.0960) acc 68.7500 (76.3281) lr 0.260000 -FPS@all 820.956, TIME@all 0.312 -epoch: [12/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:28:10 loss 2.2149 (2.0055) acc 81.2500 (79.8438) lr 0.260000 -epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.3125 (2.0941) acc 68.7500 (76.2500) lr 0.260000 -FPS@all 820.968, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:28:10 loss 2.4423 (1.9677) acc 71.8750 (79.3750) lr 0.260000 -epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.2852 (2.1149) acc 71.8750 (75.5469) lr 0.260000 -FPS@all 821.024, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:28:10 loss 2.4935 (1.9875) acc 71.8750 (80.3125) lr 0.260000 -epoch: [12/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.5237 (2.1232) acc 62.5000 (76.3281) lr 0.260000 -FPS@all 820.987, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:28:10 loss 2.2782 (1.9396) acc 71.8750 (81.4062) lr 0.260000 -epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:28:01 loss 2.0881 (2.1168) acc 78.1250 (76.1719) lr 0.260000 -FPS@all 820.981, TIME@all 0.312 -epoch: [12/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:28:10 loss 2.1213 (1.9307) acc 78.1250 (82.5000) lr 0.260000 -epoch: [12/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:28:01 loss 2.1073 (2.0569) acc 71.8750 (78.0469) lr 0.260000 -FPS@all 820.952, TIME@all 0.312 -epoch: [13/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:27:41 loss 1.9837 (1.9019) acc 78.1250 (83.7500) lr 0.260000 -epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:27:44 loss 1.9490 (1.9490) acc 84.3750 (81.2500) lr 0.260000 -FPS@all 821.521, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:27:42 loss 2.1825 (1.9398) acc 71.8750 (81.7188) lr 0.260000 -epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:27:44 loss 2.0111 (2.0049) acc 78.1250 (78.5156) lr 0.260000 -FPS@all 821.514, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:44 loss 2.4584 (1.9165) acc 71.8750 (82.6562) lr 0.260000 -epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.8755 (1.9853) acc 87.5000 (80.1562) lr 0.260000 -FPS@all 821.324, TIME@all 0.312 -epoch: [13/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:27:42 loss 2.0462 (1.9351) acc 78.1250 (80.3125) lr 0.260000 -epoch: [13/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:27:44 loss 2.2325 (1.9893) acc 71.8750 (78.9844) lr 0.260000 -FPS@all 821.473, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:42 loss 2.0369 (1.9011) acc 75.0000 (82.5000) lr 0.260000 -epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.8497 (1.9845) acc 81.2500 (80.3906) lr 0.260000 -FPS@all 821.452, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:43 loss 2.3142 (1.9082) acc 65.6250 (83.5938) lr 0.260000 -epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 1.9372 (1.9584) acc 78.1250 (81.3281) lr 0.260000 -FPS@all 821.433, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:27:43 loss 2.3263 (1.9224) acc 71.8750 (83.7500) lr 0.260000 -epoch: [13/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:27:45 loss 2.0749 (1.9442) acc 71.8750 (81.2500) lr 0.260000 -FPS@all 821.408, TIME@all 0.312 -epoch: [13/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:27:42 loss 2.2744 (1.9465) acc 75.0000 (80.6250) lr 0.260000 -epoch: [13/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 1:27:45 loss 2.0181 (1.9905) acc 71.8750 (79.0625) lr 0.260000 -FPS@all 821.437, TIME@all 0.312 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.1549 (1.7656) acc 78.1250 (86.0938) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 2.0388 (1.8583) acc 75.0000 (84.4531) lr 0.260000 -FPS@all 818.548, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:47 loss 2.1645 (1.7621) acc 65.6250 (86.0938) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:34 loss 1.8989 (1.8531) acc 87.5000 (83.4375) lr 0.260000 -FPS@all 818.604, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.0675 (1.7825) acc 84.3750 (84.8438) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:36 loss 2.3007 (1.8602) acc 78.1250 (82.9688) lr 0.260000 -FPS@all 818.478, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 1:27:49 loss 2.0988 (1.8254) acc 81.2500 (84.2188) lr 0.260000 -epoch: [14/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 1:27:36 loss 2.0766 (1.8895) acc 75.0000 (81.7969) lr 0.260000 -FPS@all 818.491, TIME@all 0.313 -epoch: [14/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2654 (1.8558) acc 71.8750 (82.6562) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 1.9649 (1.9041) acc 78.1250 (81.0156) lr 0.260000 -FPS@all 818.487, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2270 (1.8433) acc 75.0000 (82.9688) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.001 (0.006) eta 1:27:35 loss 2.2333 (1.9130) acc 68.7500 (80.9375) lr 0.260000 -FPS@all 818.520, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:49 loss 2.2180 (1.7951) acc 71.8750 (85.6250) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:27:35 loss 2.3446 (1.8823) acc 71.8750 (83.0469) lr 0.260000 -FPS@all 818.488, TIME@all 0.313 -epoch: [14/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:27:48 loss 1.9552 (1.7729) acc 78.1250 (85.9375) lr 0.260000 -epoch: [14/350][40/50] time 0.316 (0.313) data 0.001 (0.006) eta 1:27:35 loss 2.1505 (1.8777) acc 78.1250 (82.5000) lr 0.260000 -FPS@all 818.507, TIME@all 0.313 -epoch: [15/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 1:27:32 loss 1.8626 (1.7616) acc 78.1250 (86.8750) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.4890 (1.8177) acc 96.8750 (84.8438) lr 0.260000 -FPS@all 819.971, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:32 loss 1.5863 (1.7754) acc 93.7500 (88.2812) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.7300 (1.8268) acc 84.3750 (84.6875) lr 0.260000 -FPS@all 820.012, TIME@all 0.312 -epoch: [15/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 1:27:33 loss 2.0121 (1.7987) acc 81.2500 (85.6250) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.6598 (1.8514) acc 90.6250 (83.3594) lr 0.260000 -FPS@all 819.902, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.001 (0.012) eta 1:27:33 loss 1.8165 (1.7817) acc 78.1250 (86.2500) lr 0.260000 -epoch: [15/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:27:18 loss 1.8319 (1.8493) acc 84.3750 (83.9844) lr 0.260000 -FPS@all 819.937, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.011) eta 1:27:33 loss 2.0262 (1.7596) acc 81.2500 (86.2500) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.8683 (1.8221) acc 81.2500 (84.7656) lr 0.260000 -FPS@all 819.876, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:33 loss 1.9889 (1.8129) acc 75.0000 (83.5938) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:27:18 loss 1.7582 (1.8574) acc 87.5000 (82.7344) lr 0.260000 -FPS@all 819.899, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 1:27:33 loss 1.6255 (1.7841) acc 84.3750 (85.3125) lr 0.260000 -epoch: [15/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:27:18 loss 2.2408 (1.8380) acc 75.0000 (83.4375) lr 0.260000 -FPS@all 819.946, TIME@all 0.312 -epoch: [15/350][20/50] time 0.318 (0.313) data 0.001 (0.012) eta 1:27:33 loss 1.9661 (1.7990) acc 81.2500 (86.0938) lr 0.260000 -epoch: [15/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:27:18 loss 1.8645 (1.8606) acc 90.6250 (83.5938) lr 0.260000 -FPS@all 819.898, TIME@all 0.312 -epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.9626 (1.7379) acc 81.2500 (86.7188) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 2.0224 (1.7724) acc 78.1250 (85.4688) lr 0.260000 -FPS@all 821.089, TIME@all 0.312 -epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:26:58 loss 1.7966 (1.6555) acc 81.2500 (87.9688) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.9531 (1.7421) acc 71.8750 (85.7812) lr 0.260000 -FPS@all 821.115, TIME@all 0.312 -epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6974 (1.6741) acc 84.3750 (88.1250) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.8325 (1.7259) acc 84.3750 (86.7188) lr 0.260000 -FPS@all 821.033, TIME@all 0.312 -epoch: [16/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6967 (1.7068) acc 87.5000 (86.7188) lr 0.260000 -epoch: [16/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.7892 (1.7877) acc 87.5000 (84.7656) lr 0.260000 -FPS@all 821.024, TIME@all 0.312 -epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 2.0385 (1.6987) acc 78.1250 (89.0625) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 2.0935 (1.8025) acc 75.0000 (84.2188) lr 0.260000 -FPS@all 821.073, TIME@all 0.312 -epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 2.0115 (1.6608) acc 87.5000 (89.8438) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.8627 (1.7365) acc 87.5000 (87.3438) lr 0.260000 -FPS@all 821.063, TIME@all 0.312 -epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.6344 (1.6227) acc 90.6250 (90.4688) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:52 loss 1.8369 (1.7179) acc 81.2500 (87.1094) lr 0.260000 -FPS@all 821.047, TIME@all 0.312 -epoch: [16/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:26:59 loss 1.9252 (1.6842) acc 84.3750 (88.7500) lr 0.260000 -epoch: [16/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:26:53 loss 1.9735 (1.7510) acc 75.0000 (86.9531) lr 0.260000 -FPS@all 821.036, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:26:55 loss 1.7763 (1.6445) acc 84.3750 (89.8438) lr 0.260000 -epoch: [17/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:26:47 loss 1.8996 (1.6952) acc 78.1250 (87.5781) lr 0.260000 -FPS@all 820.376, TIME@all 0.312 -epoch: [17/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:26:56 loss 1.8223 (1.6305) acc 75.0000 (89.2188) lr 0.260000 -epoch: [17/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:26:48 loss 1.8678 (1.6984) acc 81.2500 (87.1875) lr 0.260000 -FPS@all 820.305, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.7350 (1.6289) acc 81.2500 (89.5312) lr 0.260000 -epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.8668 (1.6902) acc 84.3750 (87.6562) lr 0.260000 -FPS@all 820.253, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.8903 (1.6255) acc 84.3750 (89.6875) lr 0.260000 -epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.8793 (1.6891) acc 78.1250 (88.2031) lr 0.260000 -FPS@all 820.244, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:57 loss 1.7764 (1.6301) acc 87.5000 (90.6250) lr 0.260000 -epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.9442 (1.6842) acc 75.0000 (88.3594) lr 0.260000 -FPS@all 820.269, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:57 loss 1.6899 (1.6538) acc 87.5000 (89.5312) lr 0.260000 -epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.5833 (1.7040) acc 96.8750 (88.0469) lr 0.260000 -FPS@all 820.281, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:26:57 loss 1.6075 (1.6221) acc 87.5000 (89.5312) lr 0.260000 -epoch: [17/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.7865 (1.6844) acc 84.3750 (87.9688) lr 0.260000 -FPS@all 820.276, TIME@all 0.312 -epoch: [17/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:26:56 loss 1.4329 (1.6028) acc 96.8750 (91.8750) lr 0.260000 -epoch: [17/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:26:48 loss 1.7390 (1.6717) acc 87.5000 (89.7656) lr 0.260000 -FPS@all 820.282, TIME@all 0.312 -epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:24 loss 1.6715 (1.5873) acc 90.6250 (90.6250) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6752 (1.6425) acc 84.3750 (88.6719) lr 0.260000 -FPS@all 821.144, TIME@all 0.312 -epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.5557 (1.6220) acc 90.6250 (89.6875) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6507 (1.6201) acc 90.6250 (90.7031) lr 0.260000 -FPS@all 821.050, TIME@all 0.312 -epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.6027 (1.5640) acc 90.6250 (91.2500) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:24 loss 1.5662 (1.6056) acc 90.6250 (90.0781) lr 0.260000 -FPS@all 820.961, TIME@all 0.312 -epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.6935 (1.5632) acc 90.6250 (92.6562) lr 0.260000 -epoch: [18/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.6172 (1.6251) acc 93.7500 (90.7031) lr 0.260000 -FPS@all 821.026, TIME@all 0.312 -epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:26:24 loss 1.8983 (1.5618) acc 81.2500 (91.7188) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.5856 (1.5760) acc 93.7500 (90.8594) lr 0.260000 -FPS@all 821.064, TIME@all 0.312 -epoch: [18/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:26:25 loss 1.5269 (1.5832) acc 90.6250 (91.5625) lr 0.260000 -epoch: [18/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:26:23 loss 1.5916 (1.5909) acc 90.6250 (91.0156) lr 0.260000 -FPS@all 821.000, TIME@all 0.312 -epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:26:24 loss 1.6521 (1.5482) acc 90.6250 (91.8750) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.001 (0.007) eta 1:26:23 loss 1.4857 (1.5833) acc 93.7500 (90.4688) lr 0.260000 -FPS@all 821.040, TIME@all 0.312 -epoch: [18/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:26:24 loss 1.7063 (1.5715) acc 90.6250 (90.9375) lr 0.260000 -epoch: [18/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:26:23 loss 1.5518 (1.5956) acc 87.5000 (90.0000) lr 0.260000 -FPS@all 821.079, TIME@all 0.312 -epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:09 loss 2.3830 (1.6386) acc 65.6250 (88.7500) lr 0.260000 -epoch: [19/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:26:07 loss 1.5716 (1.6989) acc 87.5000 (86.9531) lr 0.260000 -FPS@all 821.271, TIME@all 0.312 -epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.8563 (1.5330) acc 84.3750 (92.9688) lr 0.260000 -epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.4016 (1.5983) acc 100.0000 (91.0156) lr 0.260000 -FPS@all 821.255, TIME@all 0.312 -epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:09 loss 1.5695 (1.6285) acc 93.7500 (90.1562) lr 0.260000 -epoch: [19/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.5778 (1.6661) acc 90.6250 (88.3594) lr 0.260000 -FPS@all 821.136, TIME@all 0.312 -epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.5845 (1.5691) acc 93.7500 (91.5625) lr 0.260000 -epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.6185 (1.6599) acc 90.6250 (88.5938) lr 0.260000 -FPS@all 821.141, TIME@all 0.312 -epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.9166 (1.6418) acc 81.2500 (89.5312) lr 0.260000 -epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.4058 (1.6638) acc 100.0000 (88.3594) lr 0.260000 -FPS@all 821.146, TIME@all 0.312 -epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:26:10 loss 1.6911 (1.5734) acc 87.5000 (91.5625) lr 0.260000 -epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:26:09 loss 1.6064 (1.6249) acc 90.6250 (89.8438) lr 0.260000 -FPS@all 821.071, TIME@all 0.312 -epoch: [19/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.7532 (1.5856) acc 81.2500 (90.6250) lr 0.260000 -epoch: [19/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:26:08 loss 1.8280 (1.6303) acc 93.7500 (90.0000) lr 0.260000 -FPS@all 821.117, TIME@all 0.312 -epoch: [19/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:26:10 loss 1.8908 (1.5553) acc 81.2500 (92.6562) lr 0.260000 -epoch: [19/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:26:09 loss 1.6892 (1.6409) acc 87.5000 (89.5312) lr 0.260000 -FPS@all 821.165, TIME@all 0.312 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:25:41 loss 1.7847 (1.5536) acc 84.3750 (90.3125) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.007) eta 1:25:41 loss 1.7815 (1.6190) acc 87.5000 (88.7500) lr 0.260000 -FPS@all 823.080, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.6803 (1.5365) acc 87.5000 (91.2500) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.8391 (1.5954) acc 87.5000 (90.2344) lr 0.260000 -FPS@all 823.147, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:42 loss 1.6136 (1.5561) acc 84.3750 (91.4062) lr 0.260000 -epoch: [20/350][40/50] time 0.309 (0.311) data 0.000 (0.006) eta 1:25:42 loss 1.5522 (1.6205) acc 90.6250 (90.0781) lr 0.260000 -FPS@all 822.994, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:42 loss 1.6710 (1.5545) acc 93.7500 (92.5000) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:42 loss 1.6654 (1.6119) acc 87.5000 (90.0781) lr 0.260000 -FPS@all 823.016, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.7364 (1.6071) acc 87.5000 (89.8438) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.5424 (1.6320) acc 87.5000 (89.2188) lr 0.260000 -FPS@all 823.022, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:25:41 loss 1.6891 (1.5813) acc 87.5000 (90.7812) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.7291 (1.6304) acc 84.3750 (89.3750) lr 0.260000 -FPS@all 823.037, TIME@all 0.311 -epoch: [20/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:25:43 loss 1.6846 (1.5600) acc 84.3750 (92.0312) lr 0.260000 -epoch: [20/350][40/50] time 0.310 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.6395 (1.6013) acc 90.6250 (90.8594) lr 0.260000 -FPS@all 823.048, TIME@all 0.311 -epoch: [20/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:25:41 loss 1.5817 (1.5890) acc 90.6250 (91.4062) lr 0.260000 -epoch: [20/350][40/50] time 0.309 (0.311) data 0.000 (0.006) eta 1:25:41 loss 1.4465 (1.6364) acc 90.6250 (89.1406) lr 0.260000 -FPS@all 823.023, TIME@all 0.311 -epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.7868 (1.5410) acc 87.5000 (91.8750) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:51 loss 1.5713 (1.5955) acc 93.7500 (90.6250) lr 0.260000 -FPS@all 819.334, TIME@all 0.312 -epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.7525 (1.5567) acc 81.2500 (91.7188) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:52 loss 1.6842 (1.5877) acc 81.2500 (90.4688) lr 0.260000 -FPS@all 819.247, TIME@all 0.312 -epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.5501 (1.5733) acc 90.6250 (89.3750) lr 0.260000 -epoch: [21/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.8139 (1.6260) acc 81.2500 (88.1250) lr 0.260000 -FPS@all 819.206, TIME@all 0.312 -epoch: [21/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:25:50 loss 1.5850 (1.5124) acc 93.7500 (93.4375) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.7471 (1.5828) acc 84.3750 (91.7188) lr 0.260000 -FPS@all 819.188, TIME@all 0.313 -epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.5787 (1.5192) acc 90.6250 (93.7500) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:52 loss 1.5839 (1.5715) acc 90.6250 (91.9531) lr 0.260000 -FPS@all 819.215, TIME@all 0.312 -epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.6615 (1.5549) acc 90.6250 (92.9688) lr 0.260000 -epoch: [21/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:52 loss 1.5989 (1.6092) acc 93.7500 (91.0938) lr 0.260000 -FPS@all 819.231, TIME@all 0.312 -epoch: [21/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:49 loss 1.7835 (1.5895) acc 84.3750 (91.0938) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:25:52 loss 1.8190 (1.6289) acc 81.2500 (89.3750) lr 0.260000 -FPS@all 819.220, TIME@all 0.312 -epoch: [21/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:49 loss 1.4475 (1.5246) acc 96.8750 (93.1250) lr 0.260000 -epoch: [21/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:52 loss 1.5312 (1.5960) acc 96.8750 (90.4688) lr 0.260000 -FPS@all 819.230, TIME@all 0.312 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:26 loss 1.5565 (1.5155) acc 93.7500 (92.5000) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.6679 (1.5861) acc 90.6250 (90.4688) lr 0.260000 -FPS@all 822.416, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:26 loss 1.7279 (1.4898) acc 87.5000 (94.2188) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:13 loss 1.6047 (1.5567) acc 90.6250 (91.7188) lr 0.260000 -FPS@all 822.483, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:25:28 loss 1.6697 (1.5038) acc 75.0000 (92.9688) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:25:14 loss 1.6614 (1.5683) acc 87.5000 (91.8750) lr 0.260000 -FPS@all 822.309, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.6048 (1.4859) acc 93.7500 (93.7500) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.4808 (1.5697) acc 93.7500 (91.4062) lr 0.260000 -FPS@all 822.360, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.5731 (1.5341) acc 90.6250 (93.5938) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:25:14 loss 1.6963 (1.5518) acc 84.3750 (92.1094) lr 0.260000 -FPS@all 822.343, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:27 loss 1.5886 (1.5144) acc 93.7500 (92.6562) lr 0.260000 -epoch: [22/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.6312 (1.5789) acc 87.5000 (91.0938) lr 0.260000 -FPS@all 822.345, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.5611 (1.5027) acc 84.3750 (92.5000) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.7899 (1.5576) acc 81.2500 (91.3281) lr 0.260000 -FPS@all 822.277, TIME@all 0.311 -epoch: [22/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:25:28 loss 1.4891 (1.5125) acc 96.8750 (93.7500) lr 0.260000 -epoch: [22/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:25:14 loss 1.5369 (1.5716) acc 96.8750 (91.7188) lr 0.260000 -FPS@all 822.345, TIME@all 0.311 -epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:25:17 loss 1.7702 (1.5403) acc 87.5000 (91.2500) lr 0.260000 -epoch: [23/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:21 loss 1.5015 (1.5722) acc 90.6250 (90.6250) lr 0.260000 -FPS@all 819.361, TIME@all 0.312 -epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:25:18 loss 1.6664 (1.5323) acc 87.5000 (92.6562) lr 0.260000 -epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4780 (1.5620) acc 90.6250 (91.4062) lr 0.260000 -FPS@all 819.254, TIME@all 0.312 -epoch: [23/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:25:18 loss 1.5097 (1.5438) acc 90.6250 (90.9375) lr 0.260000 -epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:22 loss 1.4420 (1.5536) acc 90.6250 (91.0938) lr 0.260000 -FPS@all 819.160, TIME@all 0.313 -epoch: [23/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:25:19 loss 1.7214 (1.5280) acc 87.5000 (91.7188) lr 0.260000 -epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:22 loss 1.6131 (1.5598) acc 90.6250 (90.5469) lr 0.260000 -FPS@all 819.181, TIME@all 0.313 -epoch: [23/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:25:18 loss 1.4078 (1.5490) acc 96.8750 (91.4062) lr 0.260000 -epoch: [23/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4210 (1.5597) acc 96.8750 (91.7969) lr 0.260000 -FPS@all 819.304, TIME@all 0.312 -epoch: [23/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:25:18 loss 1.6521 (1.5040) acc 90.6250 (93.1250) lr 0.260000 -epoch: [23/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:22 loss 1.4713 (1.5296) acc 96.8750 (91.7188) lr 0.260000 -FPS@all 819.207, TIME@all 0.312 -epoch: [23/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:25:18 loss 1.4584 (1.5363) acc 93.7500 (92.6562) lr 0.260000 -epoch: [23/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:25:22 loss 1.4382 (1.5483) acc 93.7500 (92.1875) lr 0.260000 -FPS@all 819.208, TIME@all 0.312 -epoch: [23/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:25:18 loss 1.6535 (1.4719) acc 84.3750 (94.3750) lr 0.260000 -epoch: [23/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:22 loss 1.6647 (1.5356) acc 84.3750 (92.5000) lr 0.260000 -FPS@all 819.261, TIME@all 0.312 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4750 (1.4499) acc 93.7500 (94.8438) lr 0.260000 -epoch: [24/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:25:08 loss 1.7962 (1.5007) acc 75.0000 (92.9688) lr 0.260000 -FPS@all 818.933, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.5692 (1.4956) acc 87.5000 (92.8125) lr 0.260000 -epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.5767 (1.4912) acc 90.6250 (93.2812) lr 0.260000 -FPS@all 818.874, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:25:06 loss 1.3935 (1.4988) acc 93.7500 (94.5312) lr 0.260000 -epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:25:09 loss 1.6542 (1.5023) acc 87.5000 (93.8281) lr 0.260000 -FPS@all 818.844, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:07 loss 1.5128 (1.4872) acc 93.7500 (93.4375) lr 0.260000 -epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.4863 (1.5115) acc 93.7500 (93.1250) lr 0.260000 -FPS@all 818.728, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4324 (1.4779) acc 93.7500 (94.0625) lr 0.260000 -epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.3412 (1.5241) acc 96.8750 (92.8906) lr 0.260000 -FPS@all 818.815, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:25:07 loss 1.3886 (1.5098) acc 96.8750 (92.9688) lr 0.260000 -epoch: [24/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:25:09 loss 1.6375 (1.5465) acc 93.7500 (92.6562) lr 0.260000 -FPS@all 818.733, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.4676 (1.4700) acc 93.7500 (95.3125) lr 0.260000 -epoch: [24/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.8077 (1.5185) acc 84.3750 (92.7344) lr 0.260000 -FPS@all 818.784, TIME@all 0.313 -epoch: [24/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:25:06 loss 1.5346 (1.4835) acc 93.7500 (94.2188) lr 0.260000 -epoch: [24/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:25:09 loss 1.5902 (1.5221) acc 90.6250 (92.5781) lr 0.260000 -FPS@all 818.801, TIME@all 0.313 -epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:24:36 loss 1.6575 (1.5398) acc 93.7500 (91.2500) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:24:35 loss 1.4669 (1.5589) acc 90.6250 (90.0781) lr 0.260000 -FPS@all 821.650, TIME@all 0.312 -epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:24:38 loss 1.6323 (1.5345) acc 90.6250 (93.7500) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.5432 (1.5545) acc 90.6250 (91.9531) lr 0.260000 -FPS@all 821.533, TIME@all 0.312 -epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:24:37 loss 1.5416 (1.4944) acc 90.6250 (92.6562) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:24:36 loss 1.3653 (1.5414) acc 100.0000 (91.3281) lr 0.260000 -FPS@all 821.570, TIME@all 0.312 -epoch: [25/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:24:38 loss 1.5862 (1.5524) acc 93.7500 (90.1562) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4607 (1.5549) acc 93.7500 (91.4844) lr 0.260000 -FPS@all 821.489, TIME@all 0.312 -epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.5154 (1.4999) acc 90.6250 (93.2812) lr 0.260000 -epoch: [25/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4267 (1.5069) acc 90.6250 (93.2031) lr 0.260000 -FPS@all 821.524, TIME@all 0.312 -epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.5492 (1.4890) acc 87.5000 (93.7500) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4962 (1.5225) acc 93.7500 (92.5781) lr 0.260000 -FPS@all 821.550, TIME@all 0.312 -epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:37 loss 1.4934 (1.4866) acc 93.7500 (93.9062) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:36 loss 1.4729 (1.5239) acc 87.5000 (92.4219) lr 0.260000 -FPS@all 821.565, TIME@all 0.312 -epoch: [25/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:24:36 loss 1.6209 (1.5181) acc 87.5000 (92.3438) lr 0.260000 -epoch: [25/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:24:35 loss 1.5020 (1.5536) acc 93.7500 (91.4062) lr 0.260000 -FPS@all 821.589, TIME@all 0.312 -epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.5639 (1.4421) acc 90.6250 (94.0625) lr 0.260000 -epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.4490 (1.5070) acc 96.8750 (92.1094) lr 0.260000 -FPS@all 819.451, TIME@all 0.312 -epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:40 loss 1.4258 (1.4228) acc 96.8750 (95.7812) lr 0.260000 -epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:31 loss 1.4979 (1.4618) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 819.304, TIME@all 0.312 -epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:38 loss 1.4228 (1.4274) acc 96.8750 (94.3750) lr 0.260000 -epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.5030 (1.4775) acc 90.6250 (93.3594) lr 0.260000 -FPS@all 819.418, TIME@all 0.312 -epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:24:39 loss 1.5766 (1.4439) acc 90.6250 (93.7500) lr 0.260000 -epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:24:31 loss 1.4421 (1.4967) acc 96.8750 (92.9688) lr 0.260000 -FPS@all 819.304, TIME@all 0.312 -epoch: [26/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.4220 (1.4260) acc 100.0000 (95.4688) lr 0.260000 -epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.5720 (1.4863) acc 90.6250 (92.9688) lr 0.260000 -FPS@all 819.365, TIME@all 0.312 -epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:39 loss 1.5034 (1.4309) acc 96.8750 (94.8438) lr 0.260000 -epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:24:30 loss 1.3989 (1.4764) acc 96.8750 (93.6719) lr 0.260000 -FPS@all 819.371, TIME@all 0.312 -epoch: [26/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:24:39 loss 1.5285 (1.4272) acc 93.7500 (95.9375) lr 0.260000 -epoch: [26/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:24:30 loss 1.5201 (1.4971) acc 96.8750 (93.6719) lr 0.260000 -FPS@all 819.381, TIME@all 0.312 -epoch: [26/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:24:40 loss 1.5282 (1.4895) acc 87.5000 (93.1250) lr 0.260000 -epoch: [26/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:24:31 loss 1.6047 (1.5262) acc 90.6250 (92.5000) lr 0.260000 -FPS@all 819.318, TIME@all 0.312 -epoch: [27/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 1:24:14 loss 1.3199 (1.4374) acc 100.0000 (94.3750) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:00 loss 1.6500 (1.4866) acc 87.5000 (93.2812) lr 0.260000 -FPS@all 821.657, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 1:24:15 loss 1.4118 (1.4190) acc 96.8750 (95.3125) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:24:01 loss 1.5913 (1.4713) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 821.489, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5150 (1.4186) acc 90.6250 (95.1562) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.5893 (1.4833) acc 84.3750 (93.4375) lr 0.260000 -FPS@all 821.460, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:14 loss 1.5717 (1.4449) acc 90.6250 (95.0000) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:00 loss 1.5136 (1.5098) acc 93.7500 (93.0469) lr 0.260000 -FPS@all 821.544, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5690 (1.4168) acc 93.7500 (95.6250) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.6238 (1.4746) acc 93.7500 (93.9844) lr 0.260000 -FPS@all 821.480, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.4096 (1.4111) acc 96.8750 (95.4688) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.5977 (1.4766) acc 90.6250 (93.4375) lr 0.260000 -FPS@all 821.510, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5878 (1.4378) acc 87.5000 (94.2188) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.4629 (1.4715) acc 93.7500 (93.5938) lr 0.260000 -FPS@all 821.515, TIME@all 0.312 -epoch: [27/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 1:24:15 loss 1.5008 (1.4471) acc 90.6250 (94.2188) lr 0.260000 -epoch: [27/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:24:01 loss 1.6759 (1.4845) acc 84.3750 (93.7500) lr 0.260000 -FPS@all 821.505, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:34 loss 1.3526 (1.4557) acc 96.8750 (94.8438) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:43 loss 1.3564 (1.4703) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 821.865, TIME@all 0.311 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:34 loss 1.3073 (1.4758) acc 100.0000 (94.5312) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:43 loss 1.5193 (1.5056) acc 90.6250 (93.0469) lr 0.260000 -FPS@all 821.779, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:35 loss 1.3095 (1.4245) acc 100.0000 (95.7812) lr 0.260000 -epoch: [28/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.3887 (1.4792) acc 90.6250 (93.6719) lr 0.260000 -FPS@all 821.713, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.011) eta 1:23:35 loss 1.4862 (1.4779) acc 93.7500 (94.2188) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.3751 (1.5002) acc 96.8750 (93.0469) lr 0.260000 -FPS@all 821.697, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:23:35 loss 1.6803 (1.4677) acc 87.5000 (93.7500) lr 0.260000 -epoch: [28/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5154 (1.4982) acc 93.7500 (93.0469) lr 0.260000 -FPS@all 821.727, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.012) eta 1:23:35 loss 1.5152 (1.4850) acc 90.6250 (93.2812) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5318 (1.5160) acc 93.7500 (92.4219) lr 0.260000 -FPS@all 821.748, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.000 (0.013) eta 1:23:35 loss 1.3699 (1.4909) acc 96.8750 (93.7500) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.5084 (1.5226) acc 93.7500 (91.8750) lr 0.260000 -FPS@all 821.711, TIME@all 0.312 -epoch: [28/350][20/50] time 0.312 (0.311) data 0.001 (0.012) eta 1:23:35 loss 1.5735 (1.4754) acc 87.5000 (94.0625) lr 0.260000 -epoch: [28/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:23:44 loss 1.4265 (1.5083) acc 100.0000 (93.6719) lr 0.260000 -FPS@all 821.757, TIME@all 0.312 -epoch: [29/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.4936 (1.4160) acc 90.6250 (95.0000) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3840 (1.4551) acc 100.0000 (94.1406) lr 0.260000 -FPS@all 820.661, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.4282 (1.4216) acc 90.6250 (95.3125) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:36 loss 1.5305 (1.4687) acc 90.6250 (92.7344) lr 0.260000 -FPS@all 820.671, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.3515 (1.4073) acc 96.8750 (94.8438) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3879 (1.4298) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 820.557, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:23:46 loss 1.3265 (1.4087) acc 100.0000 (95.0000) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:23:37 loss 1.3521 (1.4509) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 820.540, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.2630 (1.3978) acc 96.8750 (95.4688) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.4642 (1.4492) acc 96.8750 (94.0625) lr 0.260000 -FPS@all 820.561, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.4438 (1.4284) acc 93.7500 (93.7500) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3917 (1.4376) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 820.556, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:46 loss 1.4322 (1.4140) acc 96.8750 (95.0000) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.5395 (1.4669) acc 90.6250 (93.8281) lr 0.260000 -FPS@all 820.612, TIME@all 0.312 -epoch: [29/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:45 loss 1.5072 (1.4242) acc 93.7500 (94.8438) lr 0.260000 -epoch: [29/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:37 loss 1.3888 (1.4453) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 820.637, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:23:40 loss 1.3873 (1.4740) acc 93.7500 (93.4375) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.5961 (1.5236) acc 87.5000 (91.7188) lr 0.260000 -FPS@all 820.589, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.4927 (1.4637) acc 96.8750 (93.1250) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.6540 (1.4897) acc 84.3750 (92.8125) lr 0.260000 -FPS@all 820.555, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:40 loss 1.6383 (1.4638) acc 90.6250 (94.0625) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:25 loss 1.6774 (1.5285) acc 84.3750 (92.1875) lr 0.260000 -FPS@all 820.485, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:39 loss 1.5117 (1.4890) acc 93.7500 (92.9688) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:25 loss 1.4164 (1.5279) acc 93.7500 (91.6406) lr 0.260000 -FPS@all 820.457, TIME@all 0.312 -epoch: [30/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.5782 (1.4623) acc 87.5000 (95.1562) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:25 loss 1.5256 (1.5118) acc 93.7500 (92.5000) lr 0.260000 -FPS@all 820.497, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:40 loss 1.5389 (1.4561) acc 90.6250 (94.0625) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:24 loss 1.4736 (1.5317) acc 93.7500 (91.7188) lr 0.260000 -FPS@all 820.491, TIME@all 0.312 -epoch: [30/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:23:39 loss 1.4243 (1.4673) acc 93.7500 (93.2812) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.007) eta 1:23:24 loss 1.4600 (1.5348) acc 90.6250 (91.3281) lr 0.260000 -FPS@all 820.511, TIME@all 0.312 -epoch: [30/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:23:39 loss 1.4943 (1.4753) acc 96.8750 (94.0625) lr 0.260000 -epoch: [30/350][40/50] time 0.305 (0.313) data 0.000 (0.006) eta 1:23:24 loss 1.5571 (1.5056) acc 87.5000 (92.8125) lr 0.260000 -FPS@all 820.451, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4845 (1.4211) acc 93.7500 (93.9062) lr 0.260000 -epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.5522 (1.4661) acc 84.3750 (92.5781) lr 0.260000 -FPS@all 820.527, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4855 (1.4177) acc 96.8750 (95.1562) lr 0.260000 -epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:05 loss 1.4288 (1.4582) acc 100.0000 (93.6719) lr 0.260000 -FPS@all 820.590, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:23:08 loss 1.5682 (1.4101) acc 87.5000 (96.0938) lr 0.260000 -epoch: [31/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 1:23:06 loss 1.4159 (1.4545) acc 100.0000 (94.3750) lr 0.260000 -FPS@all 820.489, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 1:23:08 loss 1.4143 (1.4311) acc 96.8750 (95.0000) lr 0.260000 -epoch: [31/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.2883 (1.4577) acc 100.0000 (94.4531) lr 0.260000 -FPS@all 820.498, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.6776 (1.4463) acc 81.2500 (93.9062) lr 0.260000 -epoch: [31/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.4019 (1.4725) acc 100.0000 (92.7344) lr 0.260000 -FPS@all 820.488, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.3520 (1.4116) acc 96.8750 (96.0938) lr 0.260000 -epoch: [31/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:23:06 loss 1.3501 (1.4434) acc 96.8750 (95.0781) lr 0.260000 -FPS@all 820.498, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4087 (1.4098) acc 96.8750 (95.6250) lr 0.260000 -epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:07 loss 1.4141 (1.4671) acc 96.8750 (94.1406) lr 0.260000 -FPS@all 820.436, TIME@all 0.312 -epoch: [31/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:23:08 loss 1.4699 (1.4447) acc 93.7500 (95.4688) lr 0.260000 -epoch: [31/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:23:06 loss 1.3289 (1.4611) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 820.486, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:22:57 loss 1.6850 (1.3778) acc 84.3750 (95.3125) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:22:43 loss 1.4879 (1.4139) acc 96.8750 (94.6094) lr 0.260000 -FPS@all 821.433, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.8145 (1.4029) acc 84.3750 (95.1562) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3301 (1.4341) acc 100.0000 (94.4531) lr 0.260000 -FPS@all 821.460, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.6815 (1.3823) acc 93.7500 (96.7188) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.3002 (1.4056) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 821.337, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:22:58 loss 1.5301 (1.3848) acc 90.6250 (96.8750) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.4035 (1.4140) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 821.307, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.4965 (1.3613) acc 90.6250 (97.3438) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.4128 (1.4080) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 821.358, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.5097 (1.4131) acc 87.5000 (94.6875) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3919 (1.4353) acc 93.7500 (93.9062) lr 0.260000 -FPS@all 821.358, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:22:57 loss 1.3657 (1.3704) acc 93.7500 (95.3125) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:43 loss 1.3849 (1.4004) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 821.366, TIME@all 0.312 -epoch: [32/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:22:58 loss 1.6577 (1.3664) acc 93.7500 (96.8750) lr 0.260000 -epoch: [32/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:22:44 loss 1.4665 (1.4010) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 821.328, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:22:43 loss 1.3181 (1.4284) acc 100.0000 (95.6250) lr 0.260000 -epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.4651 (1.4819) acc 90.6250 (93.5156) lr 0.260000 -FPS@all 820.371, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:22:42 loss 1.3704 (1.4187) acc 96.8750 (95.4688) lr 0.260000 -epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.4664 (1.4792) acc 90.6250 (93.0469) lr 0.260000 -FPS@all 820.446, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:43 loss 1.6915 (1.4567) acc 90.6250 (95.0000) lr 0.260000 -epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.3825 (1.4818) acc 100.0000 (93.7500) lr 0.260000 -FPS@all 820.281, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:44 loss 1.3712 (1.4253) acc 93.7500 (95.0000) lr 0.260000 -epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:22:42 loss 1.4577 (1.4523) acc 90.6250 (94.2969) lr 0.260000 -FPS@all 820.244, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:42 loss 1.3153 (1.4308) acc 100.0000 (95.1562) lr 0.260000 -epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:22:41 loss 1.3471 (1.4584) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 820.379, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:44 loss 1.4781 (1.4555) acc 93.7500 (94.0625) lr 0.260000 -epoch: [33/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.5819 (1.4809) acc 90.6250 (93.4375) lr 0.260000 -FPS@all 820.269, TIME@all 0.312 -epoch: [33/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:22:43 loss 1.5082 (1.4165) acc 96.8750 (95.7812) lr 0.260000 -epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.4395 (1.4579) acc 96.8750 (94.4531) lr 0.260000 -FPS@all 820.283, TIME@all 0.312 -epoch: [33/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:22:44 loss 1.4600 (1.4722) acc 93.7500 (93.7500) lr 0.260000 -epoch: [33/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:22:42 loss 1.5006 (1.4922) acc 87.5000 (93.3594) lr 0.260000 -FPS@all 820.307, TIME@all 0.312 -epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3004 (1.4298) acc 96.8750 (95.0000) lr 0.260000 -epoch: [34/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:22:23 loss 1.8499 (1.4629) acc 87.5000 (93.4375) lr 0.260000 -FPS@all 820.182, TIME@all 0.312 -epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:27 loss 1.3592 (1.4146) acc 96.8750 (95.4688) lr 0.260000 -epoch: [34/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:22:23 loss 1.5750 (1.4391) acc 93.7500 (95.0000) lr 0.260000 -FPS@all 820.241, TIME@all 0.312 -epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3915 (1.4168) acc 100.0000 (95.3125) lr 0.260000 -epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.5279 (1.4492) acc 87.5000 (94.6094) lr 0.260000 -FPS@all 820.073, TIME@all 0.312 -epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.2267 (1.4572) acc 100.0000 (94.2188) lr 0.260000 -epoch: [34/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:22:24 loss 1.3501 (1.4694) acc 93.7500 (93.9844) lr 0.260000 -FPS@all 820.104, TIME@all 0.312 -epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:27 loss 1.3010 (1.4026) acc 96.8750 (96.0938) lr 0.260000 -epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.6941 (1.4517) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 820.155, TIME@all 0.312 -epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.3584 (1.4042) acc 96.8750 (95.3125) lr 0.260000 -epoch: [34/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:22:24 loss 1.4948 (1.4070) acc 90.6250 (95.1562) lr 0.260000 -FPS@all 820.117, TIME@all 0.312 -epoch: [34/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.4699 (1.4144) acc 93.7500 (95.4688) lr 0.260000 -epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.5347 (1.4448) acc 90.6250 (94.4531) lr 0.260000 -FPS@all 820.114, TIME@all 0.312 -epoch: [34/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:22:28 loss 1.4314 (1.4114) acc 90.6250 (95.1562) lr 0.260000 -epoch: [34/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:22:24 loss 1.4645 (1.4184) acc 90.6250 (94.9219) lr 0.260000 -FPS@all 820.151, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.3476 (1.4101) acc 100.0000 (95.0000) lr 0.260000 -epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:53 loss 1.3865 (1.4248) acc 96.8750 (94.6094) lr 0.260000 -FPS@all 821.718, TIME@all 0.312 -epoch: [35/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.3780 (1.4115) acc 96.8750 (95.1562) lr 0.260000 -epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:21:53 loss 1.3565 (1.4177) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 821.792, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.001 (0.013) eta 1:21:51 loss 1.4937 (1.3989) acc 90.6250 (96.4062) lr 0.260000 -epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.7045 (1.4260) acc 84.3750 (94.8438) lr 0.260000 -FPS@all 821.675, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.5457 (1.4135) acc 93.7500 (95.4688) lr 0.260000 -epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.4825 (1.4364) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 821.702, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:21:51 loss 1.3611 (1.4208) acc 96.8750 (94.3750) lr 0.260000 -epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.4827 (1.4307) acc 96.8750 (94.3750) lr 0.260000 -FPS@all 821.653, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:21:51 loss 1.4476 (1.4162) acc 96.8750 (96.0938) lr 0.260000 -epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.6511 (1.4354) acc 87.5000 (94.9219) lr 0.260000 -FPS@all 821.694, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:51 loss 1.5204 (1.4219) acc 93.7500 (94.6875) lr 0.260000 -epoch: [35/350][40/50] time 0.308 (0.312) data 0.000 (0.007) eta 1:21:54 loss 1.4591 (1.4398) acc 90.6250 (94.1406) lr 0.260000 -FPS@all 821.686, TIME@all 0.312 -epoch: [35/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 1:21:52 loss 1.5611 (1.4349) acc 90.6250 (93.5938) lr 0.260000 -epoch: [35/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 1:21:54 loss 1.5331 (1.4318) acc 90.6250 (93.9844) lr 0.260000 -FPS@all 821.688, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3962 (1.3926) acc 93.7500 (95.3125) lr 0.260000 -epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3841 (1.4200) acc 93.7500 (94.2188) lr 0.260000 -FPS@all 820.421, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3782 (1.3898) acc 93.7500 (95.3125) lr 0.260000 -epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.5343 (1.4287) acc 93.7500 (94.3750) lr 0.260000 -FPS@all 820.443, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.4774 (1.3566) acc 90.6250 (96.5625) lr 0.260000 -epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:52 loss 1.2916 (1.4113) acc 96.8750 (94.8438) lr 0.260000 -FPS@all 820.289, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:21:54 loss 1.4740 (1.3851) acc 96.8750 (95.3125) lr 0.260000 -epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.4540 (1.4247) acc 96.8750 (94.2188) lr 0.260000 -FPS@all 820.332, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.5673 (1.4176) acc 90.6250 (95.1562) lr 0.260000 -epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3714 (1.4209) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 820.365, TIME@all 0.312 -epoch: [36/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:21:55 loss 1.5204 (1.4149) acc 93.7500 (95.1562) lr 0.260000 -epoch: [36/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:21:51 loss 1.3479 (1.4381) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 820.298, TIME@all 0.312 -epoch: [36/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:21:55 loss 1.3733 (1.3760) acc 96.8750 (95.7812) lr 0.260000 -epoch: [36/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:21:51 loss 1.3131 (1.4046) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 820.331, TIME@all 0.312 -epoch: [36/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:21:54 loss 1.3527 (1.3771) acc 96.8750 (95.6250) lr 0.260000 -epoch: [36/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:21:51 loss 1.5210 (1.4249) acc 90.6250 (94.6094) lr 0.260000 -FPS@all 820.347, TIME@all 0.312 -epoch: [37/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:21:48 loss 1.2678 (1.3994) acc 100.0000 (95.6250) lr 0.260000 -epoch: [37/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 1:21:41 loss 1.4121 (1.4427) acc 90.6250 (94.3750) lr 0.260000 -FPS@all 819.629, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4702 (1.3943) acc 90.6250 (95.1562) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.4820 (1.4468) acc 93.7500 (93.5156) lr 0.260000 -FPS@all 819.499, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5343 (1.4002) acc 90.6250 (95.4688) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.3859 (1.4367) acc 93.7500 (93.9844) lr 0.260000 -FPS@all 819.510, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4441 (1.3728) acc 96.8750 (95.9375) lr 0.260000 -epoch: [37/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.2612 (1.4107) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 819.539, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.4820 (1.3657) acc 90.6250 (95.0000) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:21:41 loss 1.4381 (1.4417) acc 93.7500 (93.5938) lr 0.260000 -FPS@all 819.546, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.2983 (1.3554) acc 100.0000 (96.5625) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.6371 (1.4182) acc 90.6250 (95.0000) lr 0.260000 -FPS@all 819.593, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5020 (1.3754) acc 93.7500 (96.2500) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 1:21:41 loss 1.6395 (1.4411) acc 93.7500 (94.9219) lr 0.260000 -FPS@all 819.533, TIME@all 0.312 -epoch: [37/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:21:48 loss 1.5322 (1.4006) acc 96.8750 (96.4062) lr 0.260000 -epoch: [37/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:21:41 loss 1.3662 (1.4501) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 819.537, TIME@all 0.312 -epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.6916 (1.4097) acc 84.3750 (94.5312) lr 0.260000 -epoch: [38/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.5016 (1.4408) acc 93.7500 (94.6094) lr 0.260000 -FPS@all 821.207, TIME@all 0.312 -epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 1:21:17 loss 1.5796 (1.3905) acc 90.6250 (94.8438) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:14 loss 1.3329 (1.4465) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 821.259, TIME@all 0.312 -epoch: [38/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.5176 (1.4094) acc 93.7500 (95.4688) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:21:15 loss 1.4438 (1.4662) acc 90.6250 (93.2812) lr 0.260000 -FPS@all 821.113, TIME@all 0.312 -epoch: [38/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:21:18 loss 1.5379 (1.4251) acc 93.7500 (95.6250) lr 0.260000 -epoch: [38/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3989 (1.4589) acc 90.6250 (93.5938) lr 0.260000 -FPS@all 821.108, TIME@all 0.312 -epoch: [38/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 1:21:18 loss 1.4942 (1.3887) acc 90.6250 (95.9375) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3453 (1.4357) acc 100.0000 (94.8438) lr 0.260000 -FPS@all 821.128, TIME@all 0.312 -epoch: [38/350][20/50] time 0.309 (0.312) data 0.001 (0.012) eta 1:21:18 loss 1.5970 (1.4194) acc 90.6250 (95.4688) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:21:16 loss 1.6704 (1.4644) acc 84.3750 (94.0625) lr 0.260000 -FPS@all 821.079, TIME@all 0.312 -epoch: [38/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:21:20 loss 1.6906 (1.4336) acc 87.5000 (95.0000) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.3288 (1.4549) acc 100.0000 (94.2188) lr 0.260000 -FPS@all 821.126, TIME@all 0.312 -epoch: [38/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:21:18 loss 1.6247 (1.4269) acc 90.6250 (94.6875) lr 0.260000 -epoch: [38/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:21:15 loss 1.5127 (1.4692) acc 93.7500 (93.7500) lr 0.260000 -FPS@all 821.116, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:21:05 loss 1.7075 (1.4950) acc 84.3750 (93.9062) lr 0.260000 -epoch: [39/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 1:20:55 loss 1.6401 (1.5680) acc 84.3750 (91.6406) lr 0.260000 -FPS@all 821.602, TIME@all 0.312 -epoch: [39/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:21:04 loss 1.5221 (1.4659) acc 96.8750 (94.0625) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:55 loss 1.3779 (1.5115) acc 96.8750 (92.9688) lr 0.260000 -FPS@all 821.572, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.001 (0.012) eta 1:21:06 loss 1.6500 (1.4955) acc 90.6250 (92.8125) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.3543 (1.5335) acc 96.8750 (92.4219) lr 0.260000 -FPS@all 821.531, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:05 loss 1.5484 (1.4817) acc 90.6250 (93.4375) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.5722 (1.5379) acc 90.6250 (91.4062) lr 0.260000 -FPS@all 821.532, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.7963 (1.4872) acc 87.5000 (93.1250) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.3261 (1.5269) acc 96.8750 (92.1875) lr 0.260000 -FPS@all 821.452, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.5366 (1.4792) acc 90.6250 (94.5312) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.4726 (1.5668) acc 96.8750 (91.0938) lr 0.260000 -FPS@all 821.488, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:21:06 loss 1.6423 (1.4807) acc 90.6250 (93.4375) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:56 loss 1.5047 (1.5548) acc 84.3750 (91.0938) lr 0.260000 -FPS@all 821.515, TIME@all 0.312 -epoch: [39/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:21:06 loss 1.7129 (1.5195) acc 87.5000 (92.0312) lr 0.260000 -epoch: [39/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:56 loss 1.4877 (1.5402) acc 90.6250 (91.6406) lr 0.260000 -FPS@all 821.499, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3136 (1.4287) acc 96.8750 (94.8438) lr 0.260000 -epoch: [40/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:20:46 loss 1.5473 (1.4428) acc 87.5000 (94.7656) lr 0.260000 -FPS@all 820.286, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:53 loss 1.5038 (1.4372) acc 96.8750 (94.5312) lr 0.260000 -epoch: [40/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.3181 (1.4643) acc 96.8750 (93.8281) lr 0.260000 -FPS@all 820.091, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3836 (1.3867) acc 96.8750 (96.4062) lr 0.260000 -epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.5694 (1.4233) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 820.184, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:20:52 loss 1.4305 (1.4083) acc 93.7500 (95.7812) lr 0.260000 -epoch: [40/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:20:46 loss 1.6908 (1.4361) acc 90.6250 (95.5469) lr 0.260000 -FPS@all 820.211, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.2724 (1.4201) acc 100.0000 (96.0938) lr 0.260000 -epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.5743 (1.4472) acc 90.6250 (94.6875) lr 0.260000 -FPS@all 820.126, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:53 loss 1.4177 (1.3952) acc 96.8750 (95.3125) lr 0.260000 -epoch: [40/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:20:47 loss 1.4582 (1.4385) acc 93.7500 (94.1406) lr 0.260000 -FPS@all 820.104, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 1:20:52 loss 1.2783 (1.3927) acc 100.0000 (95.1562) lr 0.260000 -epoch: [40/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:20:47 loss 1.5651 (1.4155) acc 93.7500 (94.8438) lr 0.260000 -FPS@all 820.155, TIME@all 0.312 -epoch: [40/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:20:52 loss 1.3661 (1.4290) acc 100.0000 (94.6875) lr 0.260000 -epoch: [40/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:47 loss 1.4949 (1.4459) acc 93.7500 (93.8281) lr 0.260000 -FPS@all 820.162, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:41 loss 1.3526 (1.4491) acc 93.7500 (94.2188) lr 0.260000 -epoch: [41/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4507 (1.4588) acc 93.7500 (93.7500) lr 0.260000 -FPS@all 819.526, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:42 loss 1.3630 (1.4536) acc 100.0000 (93.4375) lr 0.260000 -epoch: [41/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4105 (1.4428) acc 96.8750 (93.8281) lr 0.260000 -FPS@all 819.398, TIME@all 0.312 -epoch: [41/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:20:42 loss 1.5720 (1.4376) acc 90.6250 (94.2188) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:37 loss 1.7688 (1.4663) acc 87.5000 (93.8281) lr 0.260000 -FPS@all 819.308, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:43 loss 1.3269 (1.4335) acc 100.0000 (93.7500) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.3750 (1.4518) acc 100.0000 (93.7500) lr 0.260000 -FPS@all 819.384, TIME@all 0.312 -epoch: [41/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 1:20:43 loss 1.3700 (1.4536) acc 93.7500 (93.1250) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:20:36 loss 1.3912 (1.4566) acc 96.8750 (93.7500) lr 0.260000 -FPS@all 819.389, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:20:43 loss 1.4491 (1.4775) acc 90.6250 (93.9062) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:36 loss 1.4302 (1.4702) acc 93.7500 (94.2969) lr 0.260000 -FPS@all 819.350, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:20:42 loss 1.6112 (1.4449) acc 93.7500 (94.8438) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:36 loss 1.4261 (1.4511) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 819.407, TIME@all 0.312 -epoch: [41/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:20:42 loss 1.6238 (1.4379) acc 87.5000 (93.9062) lr 0.260000 -epoch: [41/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:20:36 loss 1.4377 (1.4466) acc 96.8750 (94.2969) lr 0.260000 -FPS@all 819.416, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.3751 (1.3601) acc 96.8750 (96.0938) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:09 loss 1.4044 (1.3822) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 821.579, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:20:10 loss 1.4092 (1.3451) acc 96.8750 (97.3438) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:20:09 loss 1.3773 (1.3802) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.646, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4146 (1.3546) acc 96.8750 (97.3438) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3126 (1.3702) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 821.491, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:20:11 loss 1.5924 (1.3723) acc 96.8750 (96.7188) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:20:10 loss 1.3402 (1.3693) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 821.508, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4699 (1.3500) acc 96.8750 (96.8750) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3655 (1.3715) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 821.484, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:11 loss 1.3910 (1.3554) acc 93.7500 (96.7188) lr 0.260000 -epoch: [42/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3306 (1.3811) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 821.474, TIME@all 0.312 -epoch: [42/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:20:10 loss 1.4292 (1.3661) acc 96.8750 (96.7188) lr 0.260000 -epoch: [42/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:20:10 loss 1.3528 (1.3659) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 821.552, TIME@all 0.312 -epoch: [42/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:20:12 loss 1.3161 (1.3244) acc 100.0000 (96.8750) lr 0.260000 -epoch: [42/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:20:09 loss 1.3822 (1.3517) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 821.539, TIME@all 0.312 -epoch: [43/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.5720 (1.3442) acc 90.6250 (95.7812) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:06 loss 1.3959 (1.3895) acc 100.0000 (95.0781) lr 0.260000 -FPS@all 820.256, TIME@all 0.312 -epoch: [43/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:20:08 loss 1.4242 (1.3116) acc 96.8750 (97.3438) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:20:06 loss 1.3312 (1.3414) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 820.298, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.5717 (1.3016) acc 90.6250 (97.9688) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4750 (1.3509) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 820.141, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4149 (1.3027) acc 96.8750 (98.1250) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4016 (1.3672) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 820.192, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4476 (1.3231) acc 93.7500 (97.1875) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.7220 (1.3714) acc 87.5000 (95.6250) lr 0.260000 -FPS@all 820.152, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:20:09 loss 1.4069 (1.3272) acc 93.7500 (96.4062) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.4421 (1.3561) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 820.224, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:20:09 loss 1.3635 (1.2917) acc 96.8750 (98.4375) lr 0.260000 -epoch: [43/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:20:07 loss 1.3052 (1.3484) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 820.192, TIME@all 0.312 -epoch: [43/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:20:09 loss 1.4941 (1.2949) acc 90.6250 (97.6562) lr 0.260000 -epoch: [43/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:20:07 loss 1.4481 (1.3484) acc 87.5000 (96.0156) lr 0.260000 -FPS@all 820.191, TIME@all 0.312 -epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:19:41 loss 1.3156 (1.3437) acc 100.0000 (96.5625) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.6073 (1.3742) acc 90.6250 (96.0938) lr 0.260000 -FPS@all 821.111, TIME@all 0.312 -epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:19:42 loss 1.3358 (1.3525) acc 93.7500 (95.7812) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3337 (1.4017) acc 100.0000 (95.2344) lr 0.260000 -FPS@all 821.031, TIME@all 0.312 -epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:19:41 loss 1.3583 (1.3675) acc 96.8750 (96.7188) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:19:39 loss 1.3322 (1.3969) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 821.155, TIME@all 0.312 -epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:19:42 loss 1.3579 (1.3598) acc 96.8750 (96.4062) lr 0.260000 -epoch: [44/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3952 (1.3889) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 821.018, TIME@all 0.312 -epoch: [44/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:19:42 loss 1.3034 (1.3288) acc 96.8750 (97.5000) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:19:40 loss 1.3271 (1.3669) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 821.036, TIME@all 0.312 -epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:19:41 loss 1.5125 (1.3671) acc 93.7500 (96.5625) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.2448 (1.3999) acc 100.0000 (95.4688) lr 0.260000 -FPS@all 821.086, TIME@all 0.312 -epoch: [44/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:19:41 loss 1.5351 (1.3678) acc 93.7500 (95.4688) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.3193 (1.4079) acc 100.0000 (95.0000) lr 0.260000 -FPS@all 821.089, TIME@all 0.312 -epoch: [44/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:19:42 loss 1.5648 (1.3777) acc 84.3750 (95.7812) lr 0.260000 -epoch: [44/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:19:40 loss 1.4814 (1.3969) acc 93.7500 (95.0000) lr 0.260000 -FPS@all 821.077, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.3028 (1.3831) acc 93.7500 (96.0938) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4174 (1.4084) acc 93.7500 (95.2344) lr 0.260000 -FPS@all 820.384, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.5405 (1.3712) acc 93.7500 (96.0938) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:29 loss 1.5865 (1.4140) acc 90.6250 (94.9219) lr 0.260000 -FPS@all 820.419, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:40 loss 1.3798 (1.3600) acc 96.8750 (96.5625) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.5476 (1.3849) acc 90.6250 (95.7031) lr 0.260000 -FPS@all 820.256, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.6148 (1.3605) acc 90.6250 (95.1562) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4632 (1.4001) acc 87.5000 (94.4531) lr 0.260000 -FPS@all 820.325, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:19:39 loss 1.2842 (1.3618) acc 100.0000 (96.7188) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:19:30 loss 1.4112 (1.3888) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 820.287, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.4972 (1.3640) acc 93.7500 (96.7188) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4453 (1.4164) acc 90.6250 (95.0000) lr 0.260000 -FPS@all 820.319, TIME@all 0.312 -epoch: [45/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.4299 (1.3404) acc 93.7500 (96.7188) lr 0.260000 -epoch: [45/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.4271 (1.3888) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 820.309, TIME@all 0.312 -epoch: [45/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:39 loss 1.3147 (1.3426) acc 90.6250 (96.8750) lr 0.260000 -epoch: [45/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:19:30 loss 1.2999 (1.3731) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 820.342, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:19:24 loss 1.6308 (1.4030) acc 87.5000 (95.6250) lr 0.260000 -epoch: [46/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.5723 (1.4542) acc 84.3750 (93.8281) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.805, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.6638 (1.4344) acc 90.6250 (93.5938) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:06 loss 1.3694 (1.4628) acc 96.8750 (93.4375) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.866, TIME@all 0.311 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.6148 (1.4060) acc 84.3750 (95.0000) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.3798 (1.4765) acc 93.7500 (92.8906) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.718, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.5522 (1.3960) acc 87.5000 (95.3125) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.5152 (1.4353) acc 96.8750 (94.4531) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.789, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:19:24 loss 1.4964 (1.4024) acc 87.5000 (95.4688) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:07 loss 1.4005 (1.4414) acc 93.7500 (93.9062) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.795, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:25 loss 1.6603 (1.4177) acc 90.6250 (94.3750) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.4840 (1.4497) acc 87.5000 (93.6719) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.744, TIME@all 0.312 -epoch: [46/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:19:24 loss 1.5089 (1.3875) acc 90.6250 (95.1562) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:19:07 loss 1.3656 (1.4401) acc 100.0000 (93.9844) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.791, TIME@all 0.312 -epoch: [46/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:19:25 loss 1.5450 (1.4016) acc 96.8750 (95.7812) lr 0.260000 -epoch: [46/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:19:07 loss 1.4670 (1.4314) acc 93.7500 (94.8438) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.783, TIME@all 0.312 -epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:19:08 loss 1.5782 (1.3510) acc 84.3750 (96.7188) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:18:59 loss 1.3741 (1.3982) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 820.125, TIME@all 0.312 -epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.6272 (1.3306) acc 90.6250 (97.3438) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:18:59 loss 1.3923 (1.3921) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 820.176, TIME@all 0.312 -epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.4480 (1.3635) acc 93.7500 (95.6250) lr 0.260000 -epoch: [47/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.3510 (1.4005) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 820.047, TIME@all 0.312 -epoch: [47/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.5146 (1.3461) acc 90.6250 (97.0312) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.4076 (1.3924) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 820.117, TIME@all 0.312 -epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:19:09 loss 1.4206 (1.3304) acc 96.8750 (97.3438) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:19:00 loss 1.4131 (1.3667) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 820.062, TIME@all 0.312 -epoch: [47/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:19:08 loss 1.5562 (1.3626) acc 93.7500 (96.2500) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:18:59 loss 1.2557 (1.4036) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 820.111, TIME@all 0.312 -epoch: [47/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:19:10 loss 1.5224 (1.3827) acc 90.6250 (95.7812) lr 0.260000 -epoch: [47/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:19:00 loss 1.3557 (1.4102) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 820.075, TIME@all 0.312 -epoch: [47/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:19:09 loss 1.4529 (1.3412) acc 90.6250 (96.2500) lr 0.260000 -epoch: [47/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:19:00 loss 1.3458 (1.3874) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 820.034, TIME@all 0.312 -epoch: [48/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:18:51 loss 1.5871 (1.3541) acc 93.7500 (96.2500) lr 0.260000 -epoch: [48/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.5022 (1.4011) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 819.596, TIME@all 0.312 -epoch: [48/350][20/50] time 0.314 (0.313) data 0.001 (0.014) eta 1:18:50 loss 1.3973 (1.3341) acc 93.7500 (97.0312) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.3480 (1.3710) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 819.626, TIME@all 0.312 -epoch: [48/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.2933 (1.3441) acc 100.0000 (96.5625) lr 0.260000 -epoch: [48/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:18:52 loss 1.2812 (1.3953) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 819.497, TIME@all 0.312 -epoch: [48/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:18:50 loss 1.3780 (1.3599) acc 96.8750 (96.4062) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.5490 (1.4120) acc 90.6250 (94.6094) lr 0.260000 -FPS@all 819.515, TIME@all 0.312 -epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.2684 (1.3071) acc 100.0000 (97.8125) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.4388 (1.3687) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 819.521, TIME@all 0.312 -epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.3539 (1.3466) acc 100.0000 (96.7188) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.3005 (1.3813) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 819.493, TIME@all 0.312 -epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.6349 (1.3714) acc 90.6250 (95.4688) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:18:51 loss 1.4643 (1.3969) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 819.532, TIME@all 0.312 -epoch: [48/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:51 loss 1.4287 (1.3238) acc 96.8750 (97.5000) lr 0.260000 -epoch: [48/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 1:18:51 loss 1.3902 (1.3950) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 819.485, TIME@all 0.312 -epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:34 loss 1.4676 (1.3558) acc 96.8750 (96.5625) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:25 loss 1.5682 (1.3841) acc 90.6250 (95.7812) lr 0.260000 -FPS@all 820.362, TIME@all 0.312 -epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.5353 (1.3803) acc 90.6250 (96.5625) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:26 loss 1.5654 (1.4075) acc 87.5000 (94.9219) lr 0.260000 -FPS@all 820.195, TIME@all 0.312 -epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:18:35 loss 1.4409 (1.3654) acc 93.7500 (96.4062) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:18:26 loss 1.4084 (1.3881) acc 93.7500 (95.4688) lr 0.260000 -FPS@all 820.185, TIME@all 0.312 -epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:36 loss 1.4464 (1.3442) acc 90.6250 (95.6250) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:18:26 loss 1.4018 (1.3672) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 820.166, TIME@all 0.312 -epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.3185 (1.3620) acc 96.8750 (95.7812) lr 0.260000 -epoch: [49/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:18:26 loss 1.4061 (1.4047) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 820.220, TIME@all 0.312 -epoch: [49/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.4886 (1.3783) acc 93.7500 (95.7812) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 1:18:26 loss 1.6209 (1.4216) acc 84.3750 (94.2188) lr 0.260000 -FPS@all 820.204, TIME@all 0.312 -epoch: [49/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 1:18:35 loss 1.5025 (1.3552) acc 96.8750 (97.1875) lr 0.260000 -epoch: [49/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:18:26 loss 1.5574 (1.3767) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 820.222, TIME@all 0.312 -epoch: [49/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:18:35 loss 1.3717 (1.3555) acc 93.7500 (95.1562) lr 0.260000 -epoch: [49/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 1:18:26 loss 1.6093 (1.3936) acc 90.6250 (95.0781) lr 0.260000 -FPS@all 820.244, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:22 loss 1.3800 (1.3714) acc 96.8750 (96.2500) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.3501 (1.4032) acc 93.7500 (95.2344) lr 0.260000 -FPS@all 819.511, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:18:22 loss 1.3873 (1.4031) acc 96.8750 (94.8438) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:18:17 loss 1.3209 (1.4104) acc 96.8750 (94.5312) lr 0.260000 -FPS@all 819.548, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.4938 (1.3871) acc 93.7500 (96.5625) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.5903 (1.4055) acc 87.5000 (95.0781) lr 0.260000 -FPS@all 819.413, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.6864 (1.4180) acc 78.1250 (94.5312) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.3742 (1.4239) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 819.470, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.3207 (1.3637) acc 96.8750 (96.5625) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.4403 (1.4050) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 819.456, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:18:23 loss 1.4274 (1.3631) acc 93.7500 (96.5625) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.2475 (1.4009) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 819.437, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:22 loss 1.5662 (1.4190) acc 93.7500 (95.3125) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:17 loss 1.3772 (1.4233) acc 96.8750 (94.9219) lr 0.260000 -FPS@all 819.533, TIME@all 0.312 -epoch: [50/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:18:23 loss 1.4062 (1.3982) acc 96.8750 (95.1562) lr 0.260000 -epoch: [50/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:18:18 loss 1.2962 (1.3979) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 819.467, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:17:58 loss 1.3187 (1.3741) acc 100.0000 (95.0000) lr 0.260000 -epoch: [51/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:52 loss 1.4037 (1.4024) acc 90.6250 (93.8281) lr 0.260000 -FPS@all 820.915, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.3569 (1.3693) acc 93.7500 (96.0938) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.5966 (1.3887) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 820.799, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.3892 (1.3467) acc 96.8750 (96.8750) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.5123 (1.3934) acc 87.5000 (95.2344) lr 0.260000 -FPS@all 820.728, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:18:00 loss 1.5041 (1.3736) acc 87.5000 (95.9375) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4508 (1.4052) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 820.738, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4019 (1.3556) acc 93.7500 (96.8750) lr 0.260000 -epoch: [51/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 1:17:53 loss 1.3576 (1.3828) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 820.809, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:17:59 loss 1.5000 (1.3812) acc 90.6250 (95.6250) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 1:17:53 loss 1.4261 (1.4013) acc 93.7500 (94.9219) lr 0.260000 -FPS@all 820.771, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4435 (1.3980) acc 100.0000 (95.4688) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4925 (1.4054) acc 93.7500 (94.7656) lr 0.260000 -FPS@all 820.791, TIME@all 0.312 -epoch: [51/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:17:59 loss 1.4818 (1.3598) acc 93.7500 (96.2500) lr 0.260000 -epoch: [51/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:17:53 loss 1.4352 (1.3777) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 820.796, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.4987 (1.3678) acc 93.7500 (95.6250) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4741 (1.4008) acc 90.6250 (94.9219) lr 0.260000 -FPS@all 820.347, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 1:17:48 loss 1.5134 (1.3559) acc 90.6250 (95.7812) lr 0.260000 -epoch: [52/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4777 (1.3850) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 820.366, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.4166 (1.3802) acc 93.7500 (95.4688) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:42 loss 1.3828 (1.3914) acc 100.0000 (95.0000) lr 0.260000 -FPS@all 820.228, TIME@all 0.312 -epoch: [52/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.5333 (1.3776) acc 87.5000 (96.2500) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:42 loss 1.4610 (1.3874) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 820.214, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.3774 (1.3642) acc 96.8750 (96.2500) lr 0.260000 -epoch: [52/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4123 (1.3930) acc 96.8750 (95.0000) lr 0.260000 -FPS@all 820.274, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:17:48 loss 1.4541 (1.3444) acc 93.7500 (96.8750) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:17:41 loss 1.4111 (1.3733) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 820.310, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:49 loss 1.3968 (1.3486) acc 96.8750 (96.4062) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:41 loss 1.3575 (1.3754) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 820.306, TIME@all 0.312 -epoch: [52/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 1:17:48 loss 1.3021 (1.3754) acc 100.0000 (95.7812) lr 0.260000 -epoch: [52/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:17:41 loss 1.4398 (1.4069) acc 96.8750 (95.0781) lr 0.260000 -FPS@all 820.281, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:17:32 loss 1.2616 (1.3148) acc 100.0000 (97.1875) lr 0.260000 -epoch: [53/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:17:22 loss 1.4549 (1.3538) acc 90.6250 (96.1719) lr 0.260000 -FPS@all 821.172, TIME@all 0.312 -epoch: [53/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:17:31 loss 1.3494 (1.3551) acc 96.8750 (96.2500) lr 0.260000 -epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:17:22 loss 1.4122 (1.3669) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 821.243, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:32 loss 1.2466 (1.3352) acc 100.0000 (97.5000) lr 0.260000 -epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:22 loss 1.4051 (1.3527) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.132, TIME@all 0.312 -epoch: [53/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2417 (1.3178) acc 100.0000 (97.3438) lr 0.260000 -epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:23 loss 1.4925 (1.3329) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 821.093, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2761 (1.3295) acc 100.0000 (96.4062) lr 0.260000 -epoch: [53/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.5802 (1.3603) acc 87.5000 (96.0938) lr 0.260000 -FPS@all 821.197, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.3427 (1.3480) acc 96.8750 (96.7188) lr 0.260000 -epoch: [53/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:17:22 loss 1.3227 (1.3739) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 821.171, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:32 loss 1.3115 (1.3203) acc 96.8750 (96.7188) lr 0.260000 -epoch: [53/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.4330 (1.3624) acc 90.6250 (95.7031) lr 0.260000 -FPS@all 821.191, TIME@all 0.312 -epoch: [53/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:17:33 loss 1.2378 (1.3290) acc 100.0000 (96.8750) lr 0.260000 -epoch: [53/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 1:17:22 loss 1.5431 (1.3622) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 821.158, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.3656 (1.3364) acc 93.7500 (97.0312) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.4743 (1.3690) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 821.433, TIME@all 0.312 -epoch: [54/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:17:05 loss 1.3529 (1.3624) acc 100.0000 (96.0938) lr 0.260000 -epoch: [54/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.4001 (1.3907) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 821.498, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4186 (1.3545) acc 90.6250 (95.9375) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:17:02 loss 1.4172 (1.3919) acc 90.6250 (95.5469) lr 0.260000 -FPS@all 821.375, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4387 (1.3183) acc 90.6250 (97.8125) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.2952 (1.3574) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 821.410, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:17:06 loss 1.3667 (1.3277) acc 90.6250 (96.7188) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:17:02 loss 1.2975 (1.3587) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 821.383, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4305 (1.3552) acc 96.8750 (96.0938) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.3572 (1.3876) acc 90.6250 (95.2344) lr 0.260000 -FPS@all 821.414, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.3067 (1.3445) acc 100.0000 (96.7188) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:02 loss 1.2166 (1.3887) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 821.378, TIME@all 0.312 -epoch: [54/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:17:06 loss 1.4257 (1.3356) acc 93.7500 (97.6562) lr 0.260000 -epoch: [54/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:17:01 loss 1.3056 (1.3748) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 821.427, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:28 loss 1.4950 (1.3674) acc 93.7500 (96.4062) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.4399 (1.3768) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 819.501, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:17:26 loss 1.6928 (1.3991) acc 87.5000 (95.0000) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 1:17:01 loss 1.5761 (1.3802) acc 90.6250 (95.9375) lr 0.260000 -FPS@all 819.592, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5493 (1.3744) acc 87.5000 (94.8438) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.4327 (1.3879) acc 90.6250 (94.8438) lr 0.260000 -FPS@all 819.396, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.3978 (1.3819) acc 93.7500 (94.8438) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.6188 (1.4103) acc 84.3750 (93.7500) lr 0.260000 -FPS@all 819.422, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5037 (1.3745) acc 93.7500 (95.3125) lr 0.260000 -epoch: [55/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.5257 (1.3956) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 819.499, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.5265 (1.3772) acc 93.7500 (95.7812) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:17:01 loss 1.3765 (1.3875) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 819.466, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.7319 (1.3677) acc 84.3750 (95.9375) lr 0.260000 -epoch: [55/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:17:02 loss 1.5526 (1.3971) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 819.442, TIME@all 0.312 -epoch: [55/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:17:27 loss 1.4391 (1.3580) acc 93.7500 (97.1875) lr 0.260000 -epoch: [55/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:17:02 loss 1.4074 (1.3920) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 819.466, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3539 (1.3998) acc 96.8750 (94.6875) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:36 loss 1.4322 (1.4137) acc 93.7500 (94.0625) lr 0.260000 -FPS@all 820.703, TIME@all 0.312 -epoch: [56/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:16:36 loss 1.2442 (1.3675) acc 100.0000 (95.6250) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:16:36 loss 1.3242 (1.3817) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 820.788, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3510 (1.4060) acc 100.0000 (96.0938) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5345 (1.4360) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 820.671, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:36 loss 1.2942 (1.3909) acc 96.8750 (95.3125) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5237 (1.4012) acc 93.7500 (95.8594) lr 0.260000 -FPS@all 820.615, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3778 (1.4088) acc 100.0000 (94.2188) lr 0.260000 -epoch: [56/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.5411 (1.4325) acc 93.7500 (93.8281) lr 0.260000 -FPS@all 820.714, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3143 (1.3882) acc 96.8750 (96.4062) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.3789 (1.4017) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 820.641, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3195 (1.3938) acc 96.8750 (95.6250) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:37 loss 1.4133 (1.4111) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 820.597, TIME@all 0.312 -epoch: [56/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:16:37 loss 1.3093 (1.3846) acc 100.0000 (96.2500) lr 0.260000 -epoch: [56/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:36 loss 1.5012 (1.4213) acc 93.7500 (94.6094) lr 0.260000 -FPS@all 820.703, TIME@all 0.312 -epoch: [57/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:16:39 loss 1.2969 (1.3020) acc 93.7500 (97.5000) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:16:29 loss 1.3856 (1.3631) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 819.374, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:40 loss 1.3358 (1.3336) acc 96.8750 (96.2500) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.4254 (1.3565) acc 87.5000 (95.7812) lr 0.260000 -FPS@all 819.228, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:39 loss 1.2724 (1.3447) acc 100.0000 (96.8750) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:16:29 loss 1.4266 (1.3710) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 819.298, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:39 loss 1.3339 (1.3362) acc 100.0000 (96.5625) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3892 (1.3603) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 819.323, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:40 loss 1.2520 (1.3312) acc 100.0000 (97.8125) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3417 (1.3606) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 819.247, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:39 loss 1.3003 (1.3344) acc 96.8750 (96.2500) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.2387 (1.3505) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 819.266, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:39 loss 1.3837 (1.3162) acc 93.7500 (96.5625) lr 0.260000 -epoch: [57/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.3551 (1.3598) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 819.285, TIME@all 0.312 -epoch: [57/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 1:16:39 loss 1.3726 (1.3511) acc 96.8750 (96.0938) lr 0.260000 -epoch: [57/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:16:30 loss 1.4399 (1.3749) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 819.279, TIME@all 0.312 -epoch: [58/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:16:15 loss 1.4228 (1.3011) acc 93.7500 (97.5000) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.4076 (1.3190) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 820.565, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:15 loss 1.3550 (1.3135) acc 90.6250 (97.0312) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2265 (1.3365) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 820.468, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:16:16 loss 1.3048 (1.2954) acc 96.8750 (97.1875) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:16:06 loss 1.3254 (1.3234) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 820.421, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.3031 (1.3049) acc 96.8750 (97.0312) lr 0.260000 -epoch: [58/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:16:07 loss 1.4554 (1.3353) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.414, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.4405 (1.2731) acc 96.8750 (98.2812) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.3741 (1.3100) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 820.441, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:16:16 loss 1.3387 (1.3032) acc 96.8750 (97.3438) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2582 (1.3156) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 820.434, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:16:15 loss 1.4938 (1.2976) acc 90.6250 (97.5000) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2774 (1.3189) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 820.468, TIME@all 0.312 -epoch: [58/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:16:16 loss 1.3491 (1.3124) acc 93.7500 (97.0312) lr 0.260000 -epoch: [58/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:16:06 loss 1.2659 (1.3397) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 820.437, TIME@all 0.312 -epoch: [59/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:49 loss 1.4315 (1.3114) acc 96.8750 (97.1875) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:15:41 loss 1.5558 (1.3501) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 821.663, TIME@all 0.312 -epoch: [59/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:50 loss 1.3540 (1.3244) acc 93.7500 (97.0312) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 1:15:42 loss 1.4768 (1.3416) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 821.507, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:51 loss 1.5054 (1.3415) acc 90.6250 (97.1875) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:43 loss 1.3163 (1.3432) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 821.474, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.4572 (1.3354) acc 93.7500 (97.3438) lr 0.260000 -epoch: [59/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.4024 (1.3567) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 821.506, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.5054 (1.3294) acc 96.8750 (97.0312) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.2849 (1.3496) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 821.572, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:15:50 loss 1.3944 (1.3224) acc 96.8750 (97.1875) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.3475 (1.3375) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 821.515, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:15:51 loss 1.4910 (1.3231) acc 90.6250 (96.7188) lr 0.260000 -epoch: [59/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 1:15:42 loss 1.4928 (1.3446) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 821.520, TIME@all 0.312 -epoch: [59/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:15:50 loss 1.4647 (1.3047) acc 93.7500 (97.0312) lr 0.260000 -epoch: [59/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:15:42 loss 1.2883 (1.3322) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 821.575, TIME@all 0.312 -epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.3230 (1.4094) acc 96.8750 (95.7812) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5239 (1.4930) acc 87.5000 (92.5000) lr 0.260000 -FPS@all 819.908, TIME@all 0.312 -epoch: [60/350][20/50] time 0.317 (0.313) data 0.000 (0.014) eta 1:15:45 loss 1.2973 (1.4172) acc 100.0000 (93.9062) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.4208 (1.4904) acc 93.7500 (92.4219) lr 0.260000 -FPS@all 819.960, TIME@all 0.312 -epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.6219 (1.3985) acc 90.6250 (94.0625) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5348 (1.4697) acc 96.8750 (93.5938) lr 0.260000 -FPS@all 819.843, TIME@all 0.312 -epoch: [60/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:15:46 loss 1.3096 (1.4152) acc 100.0000 (95.6250) lr 0.260000 -epoch: [60/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 1:15:37 loss 1.4139 (1.4823) acc 93.7500 (93.2031) lr 0.260000 -FPS@all 819.836, TIME@all 0.312 -epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:46 loss 1.5059 (1.4003) acc 93.7500 (95.7812) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 1:15:37 loss 1.2985 (1.4768) acc 96.8750 (93.8281) lr 0.260000 -FPS@all 819.878, TIME@all 0.312 -epoch: [60/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 1:15:47 loss 1.4281 (1.3668) acc 100.0000 (96.7188) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 1:15:37 loss 1.5582 (1.4508) acc 93.7500 (93.8281) lr 0.260000 -FPS@all 819.832, TIME@all 0.312 -epoch: [60/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 1:15:46 loss 1.3592 (1.3966) acc 93.7500 (95.1562) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.5141 (1.4944) acc 96.8750 (93.2031) lr 0.260000 -FPS@all 819.899, TIME@all 0.312 -epoch: [60/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:15:46 loss 1.4152 (1.4087) acc 96.8750 (95.3125) lr 0.260000 -epoch: [60/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:15:37 loss 1.3910 (1.4835) acc 93.7500 (93.5156) lr 0.260000 -FPS@all 819.879, TIME@all 0.312 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:15:17 loss 1.3886 (1.3982) acc 93.7500 (96.0938) lr 0.260000 -epoch: [61/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:15:06 loss 1.3181 (1.4085) acc 100.0000 (95.5469) lr 0.260000 -FPS@all 822.647, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.2945 (1.3787) acc 96.8750 (95.7812) lr 0.260000 -epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.2840 (1.3854) acc 96.8750 (95.3906) lr 0.260000 -FPS@all 822.557, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3210 (1.3745) acc 100.0000 (95.4688) lr 0.260000 -epoch: [61/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.3871 (1.4055) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 822.491, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3167 (1.3628) acc 100.0000 (96.4062) lr 0.260000 -epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.3508 (1.3855) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 822.462, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.4323 (1.3670) acc 93.7500 (96.7188) lr 0.260000 -epoch: [61/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.2593 (1.3810) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 822.475, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.4000 (1.3533) acc 90.6250 (95.6250) lr 0.260000 -epoch: [61/350][40/50] time 0.309 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.3624 (1.3863) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 822.523, TIME@all 0.311 -epoch: [61/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3816 (1.3608) acc 100.0000 (96.7188) lr 0.260000 -epoch: [61/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:15:06 loss 1.3090 (1.3860) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 822.534, TIME@all 0.311 -epoch: [61/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 1:15:17 loss 1.3386 (1.3801) acc 100.0000 (96.4062) lr 0.260000 -epoch: [61/350][40/50] time 0.308 (0.312) data 0.001 (0.006) eta 1:15:06 loss 1.2915 (1.4133) acc 100.0000 (94.6875) lr 0.260000 -FPS@all 822.502, TIME@all 0.311 -epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.4854 (1.4114) acc 93.7500 (95.3125) lr 0.260000 -epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3892 (1.4241) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 820.512, TIME@all 0.312 -epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.5050 (1.3616) acc 93.7500 (97.1875) lr 0.260000 -epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3270 (1.4067) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 820.553, TIME@all 0.312 -epoch: [62/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.8750 (1.3964) acc 84.3750 (95.9375) lr 0.260000 -epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.4430 (1.4283) acc 90.6250 (94.5312) lr 0.260000 -FPS@all 820.408, TIME@all 0.312 -epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:15:08 loss 1.6982 (1.4272) acc 90.6250 (94.0625) lr 0.260000 -epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:15:03 loss 1.4117 (1.4331) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 820.373, TIME@all 0.312 -epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.7935 (1.4276) acc 87.5000 (95.3125) lr 0.260000 -epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3242 (1.4547) acc 96.8750 (93.9062) lr 0.260000 -FPS@all 820.492, TIME@all 0.312 -epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.5440 (1.4004) acc 90.6250 (95.4688) lr 0.260000 -epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.4107 (1.4412) acc 96.8750 (94.1406) lr 0.260000 -FPS@all 820.450, TIME@all 0.312 -epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.6733 (1.3547) acc 93.7500 (96.7188) lr 0.260000 -epoch: [62/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3298 (1.4043) acc 100.0000 (95.3125) lr 0.260000 -FPS@all 820.443, TIME@all 0.312 -epoch: [62/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:15:08 loss 1.6377 (1.4091) acc 87.5000 (95.1562) lr 0.260000 -epoch: [62/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:15:03 loss 1.3386 (1.4293) acc 96.8750 (94.9219) lr 0.260000 -FPS@all 820.440, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:45 loss 1.3257 (1.3702) acc 96.8750 (95.3125) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:49 loss 1.4062 (1.4044) acc 96.8750 (94.6094) lr 0.260000 -FPS@all 820.262, TIME@all 0.312 -epoch: [63/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.3629 (1.3564) acc 96.8750 (95.7812) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:49 loss 1.3849 (1.3915) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 820.203, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3458 (1.3605) acc 96.8750 (96.2500) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4538 (1.4029) acc 93.7500 (94.8438) lr 0.260000 -FPS@all 820.081, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.011) eta 1:14:48 loss 1.3289 (1.3557) acc 93.7500 (96.0938) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3478 (1.4002) acc 100.0000 (94.7656) lr 0.260000 -FPS@all 820.054, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.4744 (1.3672) acc 96.8750 (97.0312) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4213 (1.4026) acc 90.6250 (95.7812) lr 0.260000 -FPS@all 820.079, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:47 loss 1.5316 (1.3712) acc 90.6250 (95.4688) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3408 (1.4130) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 820.111, TIME@all 0.312 -epoch: [63/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3519 (1.3766) acc 96.8750 (96.4062) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.4970 (1.4243) acc 87.5000 (94.7656) lr 0.260000 -FPS@all 820.087, TIME@all 0.312 -epoch: [63/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:48 loss 1.3891 (1.3664) acc 96.8750 (96.7188) lr 0.260000 -epoch: [63/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:14:50 loss 1.3838 (1.3908) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 820.113, TIME@all 0.312 -epoch: [64/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:14:33 loss 1.3760 (1.3813) acc 96.8750 (95.3125) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3408 (1.4015) acc 96.8750 (94.8438) lr 0.260000 -FPS@all 821.066, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:33 loss 1.6249 (1.3906) acc 93.7500 (95.7812) lr 0.260000 -epoch: [64/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.2842 (1.3759) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 820.968, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:14:34 loss 1.6213 (1.3856) acc 90.6250 (96.8750) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:14:30 loss 1.2972 (1.3946) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 820.944, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.5978 (1.4006) acc 90.6250 (94.8438) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.001 (0.007) eta 1:14:30 loss 1.3743 (1.4025) acc 96.8750 (94.7656) lr 0.260000 -FPS@all 820.993, TIME@all 0.312 -epoch: [64/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.6688 (1.4140) acc 90.6250 (95.1562) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.4291 (1.4155) acc 93.7500 (94.6094) lr 0.260000 -FPS@all 820.990, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:14:33 loss 1.6619 (1.3685) acc 93.7500 (96.5625) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3428 (1.3931) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 820.983, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:34 loss 1.4863 (1.4078) acc 90.6250 (94.5312) lr 0.260000 -epoch: [64/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:14:30 loss 1.3527 (1.4146) acc 93.7500 (94.2188) lr 0.260000 -FPS@all 820.992, TIME@all 0.312 -epoch: [64/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:14:33 loss 1.5091 (1.3952) acc 96.8750 (95.4688) lr 0.260000 -epoch: [64/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 1:14:30 loss 1.3842 (1.4041) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 821.028, TIME@all 0.312 -epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.5787 (1.3227) acc 90.6250 (97.1875) lr 0.260000 -epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.3743 (1.3373) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 820.783, TIME@all 0.312 -epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.4280 (1.3282) acc 93.7500 (97.0312) lr 0.260000 -epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.3729 (1.3498) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 820.826, TIME@all 0.312 -epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:14:21 loss 1.6557 (1.3269) acc 87.5000 (97.0312) lr 0.260000 -epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3531 (1.3538) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 820.655, TIME@all 0.312 -epoch: [65/350][20/50] time 0.313 (0.312) data 0.000 (0.011) eta 1:14:20 loss 1.6503 (1.3140) acc 90.6250 (97.5000) lr 0.260000 -epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3576 (1.3399) acc 90.6250 (96.7969) lr 0.260000 -FPS@all 820.686, TIME@all 0.312 -epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.4944 (1.3125) acc 90.6250 (97.0312) lr 0.260000 -epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.4898 (1.3406) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 820.710, TIME@all 0.312 -epoch: [65/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 1:14:20 loss 1.5958 (1.3139) acc 90.6250 (97.6562) lr 0.260000 -epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:16 loss 1.4005 (1.3426) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 820.760, TIME@all 0.312 -epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.3709 (1.3300) acc 96.8750 (97.8125) lr 0.260000 -epoch: [65/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.2913 (1.3621) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 820.735, TIME@all 0.312 -epoch: [65/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:14:20 loss 1.5333 (1.3163) acc 90.6250 (97.1875) lr 0.260000 -epoch: [65/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:14:17 loss 1.3640 (1.3288) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 820.719, TIME@all 0.312 -epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:14:03 loss 1.3526 (1.3230) acc 96.8750 (96.2500) lr 0.260000 -epoch: [66/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:13:58 loss 1.3208 (1.3400) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 820.786, TIME@all 0.312 -epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.4861 (1.3059) acc 96.8750 (97.8125) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:58 loss 1.3799 (1.3316) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 820.693, TIME@all 0.312 -epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3385 (1.2865) acc 96.8750 (97.6562) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.3671 (1.3243) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 820.663, TIME@all 0.312 -epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3083 (1.2811) acc 100.0000 (97.3438) lr 0.260000 -epoch: [66/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.1961 (1.3205) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 820.623, TIME@all 0.312 -epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3332 (1.2962) acc 96.8750 (97.6562) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.2972 (1.3417) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 820.673, TIME@all 0.312 -epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:14:03 loss 1.4675 (1.2729) acc 93.7500 (98.2812) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.4023 (1.3139) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 820.631, TIME@all 0.312 -epoch: [66/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.4543 (1.2845) acc 93.7500 (97.9688) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:58 loss 1.2716 (1.3108) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 820.712, TIME@all 0.312 -epoch: [66/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:14:03 loss 1.3340 (1.2919) acc 100.0000 (97.8125) lr 0.260000 -epoch: [66/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:13:59 loss 1.3022 (1.3185) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 820.680, TIME@all 0.312 -epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.3169 (1.2971) acc 96.8750 (97.6562) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.007) eta 1:14:31 loss 1.2926 (1.3145) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 808.523, TIME@all 0.317 -epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:47 loss 1.3945 (1.2867) acc 90.6250 (97.3438) lr 0.260000 -epoch: [67/350][40/50] time 0.323 (0.316) data 0.000 (0.007) eta 1:14:30 loss 1.3082 (1.3067) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 808.602, TIME@all 0.317 -epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.5320 (1.3086) acc 90.6250 (97.3438) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.007) eta 1:14:31 loss 1.4084 (1.3465) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 808.483, TIME@all 0.317 -epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:13:48 loss 1.5711 (1.3138) acc 90.6250 (97.0312) lr 0.260000 -epoch: [67/350][40/50] time 0.325 (0.316) data 0.000 (0.006) eta 1:14:31 loss 1.3345 (1.3450) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 808.459, TIME@all 0.317 -epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4124 (1.2944) acc 90.6250 (96.8750) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.4156 (1.3483) acc 90.6250 (95.6250) lr 0.260000 -FPS@all 808.550, TIME@all 0.317 -epoch: [67/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:13:50 loss 1.4938 (1.3460) acc 93.7500 (97.0312) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.000 (0.006) eta 1:14:31 loss 1.4156 (1.3471) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 808.462, TIME@all 0.317 -epoch: [67/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4914 (1.3259) acc 90.6250 (96.7188) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.3442 (1.3558) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 808.507, TIME@all 0.317 -epoch: [67/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 1:13:48 loss 1.4994 (1.3031) acc 93.7500 (97.3438) lr 0.260000 -epoch: [67/350][40/50] time 0.324 (0.316) data 0.001 (0.007) eta 1:14:31 loss 1.3067 (1.3418) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 808.473, TIME@all 0.317 -epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 1:13:53 loss 1.3202 (1.3111) acc 93.7500 (97.0312) lr 0.260000 -epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3929 (1.3492) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 817.806, TIME@all 0.313 -epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.015) eta 1:13:54 loss 1.2785 (1.3081) acc 100.0000 (97.6562) lr 0.260000 -epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3255 (1.3416) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 817.701, TIME@all 0.313 -epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.4153 (1.3334) acc 96.8750 (97.3438) lr 0.260000 -epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:40 loss 1.6029 (1.3578) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 817.602, TIME@all 0.313 -epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:55 loss 1.3260 (1.3027) acc 96.8750 (97.3438) lr 0.260000 -epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:40 loss 1.2901 (1.3256) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 817.560, TIME@all 0.313 -epoch: [68/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.2576 (1.3241) acc 100.0000 (97.1875) lr 0.260000 -epoch: [68/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3327 (1.3721) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 817.644, TIME@all 0.313 -epoch: [68/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:13:54 loss 1.3895 (1.3108) acc 90.6250 (97.5000) lr 0.260000 -epoch: [68/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:13:39 loss 1.3920 (1.3487) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 817.658, TIME@all 0.313 -epoch: [68/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 1:13:54 loss 1.3293 (1.3118) acc 96.8750 (97.0312) lr 0.260000 -epoch: [68/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:13:38 loss 1.2968 (1.3500) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 817.829, TIME@all 0.313 -epoch: [68/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 1:13:54 loss 1.3395 (1.3010) acc 96.8750 (97.3438) lr 0.260000 -epoch: [68/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:13:39 loss 1.5232 (1.3455) acc 87.5000 (95.9375) lr 0.260000 -FPS@all 817.676, TIME@all 0.313 -epoch: [69/350][20/50] time 0.331 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.3068 (1.2778) acc 96.8750 (98.1250) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.007) eta 1:15:20 loss 1.3625 (1.3282) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 799.432, TIME@all 0.320 -epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.2824 (1.2974) acc 96.8750 (97.0312) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.007) eta 1:15:19 loss 1.3009 (1.3352) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 799.510, TIME@all 0.320 -epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:49 loss 1.2596 (1.2939) acc 100.0000 (97.5000) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.001 (0.006) eta 1:15:20 loss 1.4113 (1.3173) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 799.494, TIME@all 0.320 -epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:48 loss 1.6438 (1.3122) acc 90.6250 (97.6562) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.006) eta 1:15:19 loss 1.4067 (1.3323) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 799.535, TIME@all 0.320 -epoch: [69/350][20/50] time 0.331 (0.323) data 0.001 (0.012) eta 1:15:48 loss 1.2333 (1.2877) acc 96.8750 (97.3438) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.000 (0.006) eta 1:15:19 loss 1.3307 (1.3271) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 799.468, TIME@all 0.320 -epoch: [69/350][20/50] time 0.331 (0.323) data 0.000 (0.013) eta 1:15:48 loss 1.3899 (1.2996) acc 93.7500 (97.0312) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.322) data 0.000 (0.007) eta 1:15:20 loss 1.4748 (1.3348) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 799.361, TIME@all 0.320 -epoch: [69/350][20/50] time 0.331 (0.323) data 0.001 (0.013) eta 1:15:47 loss 1.3700 (1.3080) acc 93.7500 (96.7188) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.321) data 0.001 (0.007) eta 1:15:19 loss 1.3156 (1.3379) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 799.644, TIME@all 0.320 -epoch: [69/350][20/50] time 0.332 (0.323) data 0.000 (0.012) eta 1:15:49 loss 1.2381 (1.2858) acc 100.0000 (97.8125) lr 0.260000 -epoch: [69/350][40/50] time 0.330 (0.322) data 0.000 (0.006) eta 1:15:20 loss 1.3402 (1.3217) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 799.382, TIME@all 0.320 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.015) eta 1:13:34 loss 1.4462 (1.3021) acc 93.7500 (97.3438) lr 0.260000 -epoch: [70/350][40/50] time 0.353 (0.320) data 0.000 (0.008) eta 1:14:49 loss 1.3160 (1.3217) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 786.875, TIME@all 0.325 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 1:13:33 loss 1.3309 (1.3224) acc 96.8750 (96.8750) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3559 (1.3435) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 786.922, TIME@all 0.325 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:35 loss 1.7236 (1.3377) acc 81.2500 (96.8750) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.321) data 0.000 (0.007) eta 1:14:50 loss 1.3513 (1.3426) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 786.709, TIME@all 0.325 -epoch: [70/350][20/50] time 0.315 (0.315) data 0.000 (0.015) eta 1:13:34 loss 1.4369 (1.3108) acc 96.8750 (96.8750) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.320) data 0.000 (0.008) eta 1:14:49 loss 1.3728 (1.3327) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 786.838, TIME@all 0.325 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:35 loss 1.5045 (1.3816) acc 90.6250 (95.6250) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3218 (1.3748) acc 93.7500 (95.5469) lr 0.260000 -FPS@all 786.906, TIME@all 0.325 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 1:13:34 loss 1.3770 (1.2873) acc 90.6250 (97.9688) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.320) data 0.000 (0.007) eta 1:14:49 loss 1.2967 (1.3308) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 786.829, TIME@all 0.325 -epoch: [70/350][20/50] time 0.315 (0.315) data 0.000 (0.014) eta 1:13:34 loss 1.3902 (1.3132) acc 100.0000 (97.8125) lr 0.260000 -epoch: [70/350][40/50] time 0.352 (0.320) data 0.001 (0.007) eta 1:14:49 loss 1.3886 (1.3696) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 786.806, TIME@all 0.325 -epoch: [70/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 1:13:34 loss 1.6675 (1.3114) acc 90.6250 (97.6562) lr 0.260000 -epoch: [70/350][40/50] time 0.353 (0.320) data 0.001 (0.007) eta 1:14:48 loss 1.2903 (1.3335) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 786.938, TIME@all 0.325 -epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.3201 (1.4088) acc 100.0000 (94.5312) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.007) eta 1:14:59 loss 1.2931 (1.4206) acc 100.0000 (94.5312) lr 0.260000 -FPS@all 801.191, TIME@all 0.320 -epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.4469 (1.3671) acc 93.7500 (96.5625) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3273 (1.4005) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 801.163, TIME@all 0.320 -epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4743 (1.4013) acc 90.6250 (96.0938) lr 0.260000 -epoch: [71/350][40/50] time 0.310 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3460 (1.3991) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 801.042, TIME@all 0.320 -epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.3255 (1.4043) acc 96.8750 (95.0000) lr 0.260000 -epoch: [71/350][40/50] time 0.313 (0.322) data 0.000 (0.006) eta 1:15:01 loss 1.2859 (1.4076) acc 96.8750 (94.6094) lr 0.260000 -FPS@all 801.038, TIME@all 0.320 -epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.3779 (1.4067) acc 93.7500 (95.0000) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3019 (1.4294) acc 90.6250 (93.9062) lr 0.260000 -FPS@all 801.116, TIME@all 0.320 -epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4024 (1.3578) acc 96.8750 (95.4688) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.006) eta 1:15:00 loss 1.3468 (1.3994) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 801.079, TIME@all 0.320 -epoch: [71/350][20/50] time 0.313 (0.337) data 0.000 (0.012) eta 1:18:29 loss 1.4404 (1.3964) acc 90.6250 (94.8438) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.001 (0.006) eta 1:15:00 loss 1.3612 (1.4110) acc 93.7500 (94.4531) lr 0.260000 -FPS@all 801.105, TIME@all 0.320 -epoch: [71/350][20/50] time 0.312 (0.337) data 0.000 (0.013) eta 1:18:28 loss 1.2574 (1.3658) acc 96.8750 (96.5625) lr 0.260000 -epoch: [71/350][40/50] time 0.311 (0.322) data 0.000 (0.007) eta 1:15:00 loss 1.3688 (1.3938) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 801.004, TIME@all 0.320 -epoch: [72/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.2588 (1.3232) acc 100.0000 (96.8750) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.4749 (1.3516) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 816.886, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.5033 (1.3127) acc 93.7500 (95.7812) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:49 loss 1.3735 (1.3602) acc 96.8750 (95.3906) lr 0.260000 -FPS@all 816.961, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:39 loss 1.3612 (1.3326) acc 96.8750 (95.6250) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.4471 (1.3799) acc 96.8750 (94.6875) lr 0.260000 -FPS@all 816.799, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:39 loss 1.3819 (1.3577) acc 96.8750 (96.0938) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.2847 (1.3753) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 816.781, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:12:40 loss 1.5533 (1.3534) acc 90.6250 (96.0938) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 1:12:50 loss 1.3379 (1.3733) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 816.836, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:40 loss 1.3960 (1.3276) acc 96.8750 (97.3438) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.3034 (1.3530) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 816.819, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:39 loss 1.4077 (1.3392) acc 96.8750 (97.5000) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.2800 (1.3780) acc 100.0000 (95.5469) lr 0.260000 -FPS@all 816.813, TIME@all 0.313 -epoch: [72/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:12:40 loss 1.4823 (1.3107) acc 93.7500 (97.3438) lr 0.260000 -epoch: [72/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 1:12:50 loss 1.4194 (1.3514) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 816.837, TIME@all 0.313 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:12:33 loss 1.7350 (1.3453) acc 81.2500 (96.2500) lr 0.260000 -epoch: [73/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:12:15 loss 1.2931 (1.3663) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 819.689, TIME@all 0.312 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.4518 (1.3165) acc 96.8750 (96.8750) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:15 loss 1.3037 (1.3452) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 819.693, TIME@all 0.312 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.6234 (1.3297) acc 87.5000 (96.7188) lr 0.260000 -epoch: [73/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.3241 (1.3398) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 819.626, TIME@all 0.312 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.5076 (1.3315) acc 96.8750 (95.3125) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4127 (1.3645) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 819.640, TIME@all 0.312 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:33 loss 1.5138 (1.3457) acc 93.7500 (96.0938) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4130 (1.3592) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 819.581, TIME@all 0.312 -epoch: [73/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 1:12:36 loss 1.3803 (1.3118) acc 96.8750 (97.3438) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.4032 (1.3329) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 819.604, TIME@all 0.312 -epoch: [73/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:12:34 loss 1.4174 (1.3204) acc 93.7500 (96.7188) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:12:16 loss 1.3684 (1.3432) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 819.595, TIME@all 0.312 -epoch: [73/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:12:34 loss 1.3281 (1.3079) acc 96.8750 (96.8750) lr 0.260000 -epoch: [73/350][40/50] time 0.308 (0.313) data 0.001 (0.006) eta 1:12:16 loss 1.4075 (1.3462) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 819.614, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:12:00 loss 1.2747 (1.2753) acc 100.0000 (98.1250) lr 0.260000 -epoch: [74/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:11:55 loss 1.4681 (1.3161) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 821.081, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:11:59 loss 1.6443 (1.2969) acc 90.6250 (97.1875) lr 0.260000 -epoch: [74/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:55 loss 1.6623 (1.3288) acc 87.5000 (96.9531) lr 0.260000 -FPS@all 821.139, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.3639 (1.2826) acc 93.7500 (97.6562) lr 0.260000 -epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.6840 (1.3336) acc 87.5000 (96.4844) lr 0.260000 -FPS@all 820.945, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5387 (1.3047) acc 90.6250 (97.0312) lr 0.260000 -epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.5470 (1.3249) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 821.017, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 1:12:00 loss 1.4990 (1.3065) acc 93.7500 (96.7188) lr 0.260000 -epoch: [74/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:11:55 loss 1.8743 (1.3374) acc 84.3750 (95.7031) lr 0.260000 -FPS@all 820.999, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5047 (1.2917) acc 87.5000 (96.8750) lr 0.260000 -epoch: [74/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.5114 (1.3189) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.967, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:12:00 loss 1.5710 (1.3059) acc 90.6250 (97.1875) lr 0.260000 -epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:55 loss 1.8039 (1.3411) acc 81.2500 (96.0156) lr 0.260000 -FPS@all 821.005, TIME@all 0.312 -epoch: [74/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:12:00 loss 1.5257 (1.3174) acc 96.8750 (97.3438) lr 0.260000 -epoch: [74/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:11:55 loss 1.5912 (1.3394) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 820.996, TIME@all 0.312 -epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.3714 (1.2960) acc 90.6250 (97.1875) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:45 loss 1.3536 (1.3504) acc 90.6250 (95.7812) lr 0.260000 -FPS@all 818.576, TIME@all 0.313 -epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:11:33 loss 1.5403 (1.2937) acc 90.6250 (96.8750) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:11:45 loss 1.5710 (1.3428) acc 90.6250 (95.9375) lr 0.260000 -FPS@all 818.660, TIME@all 0.313 -epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.5398 (1.3073) acc 87.5000 (97.3438) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3798 (1.3452) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 818.527, TIME@all 0.313 -epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4471 (1.2999) acc 93.7500 (97.9688) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.4409 (1.3435) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 818.540, TIME@all 0.313 -epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.3133 (1.2976) acc 100.0000 (97.5000) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3403 (1.3209) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 818.524, TIME@all 0.313 -epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4362 (1.2806) acc 96.8750 (98.1250) lr 0.260000 -epoch: [75/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.3951 (1.3256) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 818.503, TIME@all 0.313 -epoch: [75/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4374 (1.3001) acc 93.7500 (97.1875) lr 0.260000 -epoch: [75/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.4888 (1.3408) acc 90.6250 (96.1719) lr 0.260000 -FPS@all 818.560, TIME@all 0.313 -epoch: [75/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:11:34 loss 1.4024 (1.2788) acc 93.7500 (98.9062) lr 0.260000 -epoch: [75/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:11:46 loss 1.5049 (1.3392) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 818.558, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.014) eta 1:12:07 loss 1.3519 (1.2993) acc 96.8750 (97.5000) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3004 (1.3198) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 818.349, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.014) eta 1:12:07 loss 1.4395 (1.3245) acc 90.6250 (96.5625) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.2526 (1.3416) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 818.386, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 1:12:07 loss 1.3187 (1.3020) acc 96.8750 (97.0312) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 1:11:41 loss 1.2950 (1.3354) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 818.232, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.2157 (1.3212) acc 100.0000 (97.1875) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:42 loss 1.2890 (1.3201) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 818.224, TIME@all 0.313 -epoch: [76/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.4337 (1.3021) acc 93.7500 (98.1250) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3319 (1.3292) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 818.301, TIME@all 0.313 -epoch: [76/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.3432 (1.2976) acc 100.0000 (98.2812) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3288 (1.3232) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 818.297, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 1:12:07 loss 1.3679 (1.2907) acc 96.8750 (97.6562) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.3178 (1.3210) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 818.283, TIME@all 0.313 -epoch: [76/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 1:12:07 loss 1.4550 (1.2932) acc 96.8750 (97.9688) lr 0.260000 -epoch: [76/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 1:11:41 loss 1.2901 (1.3204) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 818.261, TIME@all 0.313 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:10 loss 1.2490 (1.3265) acc 100.0000 (96.5625) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.2620 (1.3544) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 820.386, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:11:10 loss 1.2469 (1.3218) acc 100.0000 (96.0938) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:06 loss 1.3790 (1.3493) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 820.444, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:11:11 loss 1.4093 (1.3142) acc 90.6250 (97.1875) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:11:07 loss 1.3976 (1.3742) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 820.326, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:11 loss 1.2487 (1.3332) acc 100.0000 (96.0938) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5181 (1.3730) acc 93.7500 (95.1562) lr 0.260000 -FPS@all 820.316, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 1:11:11 loss 1.2383 (1.3242) acc 100.0000 (96.7188) lr 0.260000 -epoch: [77/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.3335 (1.3550) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 820.371, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:10 loss 1.2313 (1.3241) acc 100.0000 (96.4062) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5053 (1.3784) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 820.363, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.001 (0.014) eta 1:11:11 loss 1.4102 (1.3536) acc 96.8750 (95.9375) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.5342 (1.3953) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 820.345, TIME@all 0.312 -epoch: [77/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:11:11 loss 1.3066 (1.3256) acc 100.0000 (96.7188) lr 0.260000 -epoch: [77/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:11:07 loss 1.4554 (1.3576) acc 87.5000 (96.0156) lr 0.260000 -FPS@all 820.329, TIME@all 0.312 -epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4231 (1.3314) acc 93.7500 (96.8750) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:44 loss 1.2819 (1.3486) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 822.262, TIME@all 0.311 -epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3267 (1.3088) acc 96.8750 (97.9688) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3612 (1.3462) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 822.082, TIME@all 0.311 -epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3472 (1.3042) acc 93.7500 (97.1875) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3539 (1.3472) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 822.173, TIME@all 0.311 -epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.011) eta 1:10:44 loss 1.3173 (1.3172) acc 96.8750 (96.8750) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3702 (1.3355) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 822.087, TIME@all 0.311 -epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3286 (1.3147) acc 93.7500 (97.5000) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.2921 (1.3360) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 822.112, TIME@all 0.311 -epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4249 (1.3387) acc 93.7500 (97.0312) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3134 (1.3559) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 822.137, TIME@all 0.311 -epoch: [78/350][20/50] time 0.309 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.4161 (1.3064) acc 93.7500 (96.2500) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.4013 (1.3292) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 822.149, TIME@all 0.311 -epoch: [78/350][20/50] time 0.308 (0.311) data 0.000 (0.012) eta 1:10:44 loss 1.3799 (1.3239) acc 96.8750 (97.0312) lr 0.260000 -epoch: [78/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:10:45 loss 1.3503 (1.3431) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 822.164, TIME@all 0.311 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.4074 (1.2914) acc 96.8750 (98.4375) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.2485 (1.3078) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 821.302, TIME@all 0.312 -epoch: [79/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:10:35 loss 1.3860 (1.2744) acc 100.0000 (98.4375) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:10:35 loss 1.2457 (1.2990) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 821.353, TIME@all 0.312 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 1:10:34 loss 1.3618 (1.2760) acc 93.7500 (97.8125) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 1:10:34 loss 1.4623 (1.3079) acc 90.6250 (97.1875) lr 0.260000 -FPS@all 821.505, TIME@all 0.312 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.4612 (1.2922) acc 96.8750 (97.8125) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:36 loss 1.5459 (1.3008) acc 90.6250 (97.1875) lr 0.260000 -FPS@all 821.182, TIME@all 0.312 -epoch: [79/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.2675 (1.2922) acc 96.8750 (97.9688) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.2127 (1.3055) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 821.308, TIME@all 0.312 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:35 loss 1.2224 (1.2811) acc 96.8750 (97.9688) lr 0.260000 -epoch: [79/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.3211 (1.3100) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 821.292, TIME@all 0.312 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:35 loss 1.4944 (1.3108) acc 87.5000 (96.5625) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 1:10:35 loss 1.3301 (1.3139) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.233, TIME@all 0.312 -epoch: [79/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 1:10:34 loss 1.4648 (1.2955) acc 96.8750 (96.7188) lr 0.260000 -epoch: [79/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:10:35 loss 1.4429 (1.3061) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 821.299, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.2672 (1.3116) acc 93.7500 (97.0312) lr 0.260000 -epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 1:10:27 loss 1.3655 (1.3390) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 819.871, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.2460 (1.2896) acc 100.0000 (97.9688) lr 0.260000 -epoch: [80/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:10:26 loss 1.2937 (1.3242) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 819.970, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:33 loss 1.2015 (1.2893) acc 100.0000 (98.2812) lr 0.260000 -epoch: [80/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2382 (1.3354) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 819.796, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:34 loss 1.2194 (1.2836) acc 100.0000 (98.1250) lr 0.260000 -epoch: [80/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2275 (1.3211) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 819.820, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:34 loss 1.2786 (1.3173) acc 96.8750 (96.4062) lr 0.260000 -epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.3678 (1.3455) acc 90.6250 (95.8594) lr 0.260000 -FPS@all 819.817, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:10:33 loss 1.3848 (1.3076) acc 96.8750 (97.3438) lr 0.260000 -epoch: [80/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:10:27 loss 1.2837 (1.3528) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 819.876, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:33 loss 1.4823 (1.3303) acc 93.7500 (95.7812) lr 0.260000 -epoch: [80/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 1:10:26 loss 1.2121 (1.3533) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 819.862, TIME@all 0.312 -epoch: [80/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:10:34 loss 1.3506 (1.3001) acc 96.8750 (97.9688) lr 0.260000 -epoch: [80/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 1:10:27 loss 1.3188 (1.3437) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 819.778, TIME@all 0.312 -epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:06 loss 1.1982 (1.2807) acc 100.0000 (97.3438) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2367 (1.2995) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 821.662, TIME@all 0.312 -epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:07 loss 1.4024 (1.2814) acc 93.7500 (98.2812) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:10:02 loss 1.2420 (1.2847) acc 96.8750 (97.8125) lr 0.260000 -FPS@all 821.723, TIME@all 0.312 -epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2861 (1.2762) acc 93.7500 (97.3438) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:03 loss 1.2127 (1.2891) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 821.581, TIME@all 0.312 -epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2213 (1.2700) acc 100.0000 (97.9688) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:03 loss 1.3114 (1.3037) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 821.537, TIME@all 0.312 -epoch: [81/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2205 (1.2854) acc 100.0000 (97.9688) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.3295 (1.3011) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 821.618, TIME@all 0.312 -epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2147 (1.2601) acc 100.0000 (99.3750) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2580 (1.2909) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 821.643, TIME@all 0.312 -epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:10:07 loss 1.2589 (1.2899) acc 100.0000 (97.3438) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2248 (1.2886) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 821.625, TIME@all 0.312 -epoch: [81/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:10:07 loss 1.3834 (1.3068) acc 93.7500 (97.1875) lr 0.260000 -epoch: [81/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:10:02 loss 1.2640 (1.3130) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 821.611, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 1:09:49 loss 1.2833 (1.2481) acc 93.7500 (98.7500) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:46 loss 1.5198 (1.3024) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 821.303, TIME@all 0.312 -epoch: [82/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:09:49 loss 1.4863 (1.2737) acc 87.5000 (97.6562) lr 0.260000 -epoch: [82/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3204 (1.2986) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 821.200, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.2881 (1.2668) acc 96.8750 (97.9688) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.2417 (1.3095) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 821.113, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.4419 (1.2727) acc 93.7500 (98.2812) lr 0.260000 -epoch: [82/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3027 (1.3045) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 821.120, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.2680 (1.2725) acc 100.0000 (97.5000) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3733 (1.3107) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 821.147, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.3142 (1.2730) acc 96.8750 (97.9688) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3324 (1.3048) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.140, TIME@all 0.312 -epoch: [82/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:09:50 loss 1.3482 (1.2613) acc 93.7500 (97.9688) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3518 (1.2861) acc 93.7500 (97.9688) lr 0.260000 -FPS@all 821.182, TIME@all 0.312 -epoch: [82/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:09:49 loss 1.2960 (1.2503) acc 100.0000 (98.9062) lr 0.260000 -epoch: [82/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:09:47 loss 1.3333 (1.2965) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 821.191, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 1:09:43 loss 1.3687 (1.2880) acc 96.8750 (97.5000) lr 0.260000 -epoch: [83/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.3875 (1.3137) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 820.042, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:09:43 loss 1.3098 (1.2762) acc 100.0000 (98.5938) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2236 (1.3127) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 819.990, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.3274 (1.2574) acc 96.8750 (98.4375) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:36 loss 1.3213 (1.2887) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 819.890, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.3876 (1.2802) acc 96.8750 (97.6562) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.3014 (1.3084) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 819.949, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:09:43 loss 1.3107 (1.3000) acc 93.7500 (97.5000) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.2524 (1.3201) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 819.926, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:09:42 loss 1.2625 (1.2624) acc 96.8750 (97.8125) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2708 (1.3086) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 820.028, TIME@all 0.312 -epoch: [83/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 1:09:43 loss 1.2673 (1.2688) acc 96.8750 (97.9688) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:09:35 loss 1.2945 (1.3013) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 819.991, TIME@all 0.312 -epoch: [83/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:42 loss 1.2977 (1.2840) acc 100.0000 (98.4375) lr 0.260000 -epoch: [83/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:09:35 loss 1.2728 (1.3013) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 819.985, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:31 loss 1.4490 (1.3465) acc 100.0000 (97.3438) lr 0.260000 -epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.2672 (1.3798) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 820.324, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:29 loss 1.5324 (1.3700) acc 96.8750 (96.2500) lr 0.260000 -epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.4581 (1.4079) acc 87.5000 (95.2344) lr 0.260000 -FPS@all 820.572, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:30 loss 1.6230 (1.4279) acc 90.6250 (94.5312) lr 0.260000 -epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.2552 (1.4352) acc 100.0000 (94.3750) lr 0.260000 -FPS@all 820.434, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.5198 (1.3871) acc 93.7500 (95.7812) lr 0.260000 -epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2925 (1.4278) acc 100.0000 (94.3750) lr 0.260000 -FPS@all 820.387, TIME@all 0.312 -epoch: [84/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.3933 (1.3610) acc 96.8750 (95.9375) lr 0.260000 -epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2947 (1.3928) acc 96.8750 (95.3125) lr 0.260000 -FPS@all 820.440, TIME@all 0.312 -epoch: [84/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.3530 (1.3673) acc 96.8750 (95.3125) lr 0.260000 -epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.2095 (1.4072) acc 100.0000 (94.6094) lr 0.260000 -FPS@all 820.416, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:09:30 loss 1.4328 (1.3675) acc 93.7500 (96.0938) lr 0.260000 -epoch: [84/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:09:24 loss 1.5120 (1.3997) acc 90.6250 (95.7031) lr 0.260000 -FPS@all 820.394, TIME@all 0.312 -epoch: [84/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:09:31 loss 1.5651 (1.4008) acc 93.7500 (94.8438) lr 0.260000 -epoch: [84/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:09:24 loss 1.5685 (1.4197) acc 87.5000 (94.6094) lr 0.260000 -FPS@all 820.365, TIME@all 0.312 -epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6326 (1.3889) acc 87.5000 (95.6250) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:55 loss 1.5579 (1.4250) acc 87.5000 (94.6875) lr 0.260000 -FPS@all 822.116, TIME@all 0.311 -epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.5935 (1.3863) acc 87.5000 (95.0000) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.3696 (1.3996) acc 93.7500 (94.5312) lr 0.260000 -FPS@all 822.050, TIME@all 0.311 -epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:09:15 loss 1.5326 (1.3998) acc 87.5000 (94.3750) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4394 (1.4085) acc 90.6250 (94.3750) lr 0.260000 -FPS@all 822.015, TIME@all 0.311 -epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6116 (1.3903) acc 93.7500 (95.7812) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4686 (1.4062) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 822.000, TIME@all 0.311 -epoch: [85/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.5496 (1.3910) acc 90.6250 (96.0938) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.3804 (1.3975) acc 96.8750 (95.6250) lr 0.260000 -FPS@all 822.042, TIME@all 0.311 -epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.6175 (1.4014) acc 93.7500 (94.6875) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4397 (1.4056) acc 87.5000 (94.6875) lr 0.260000 -FPS@all 822.044, TIME@all 0.311 -epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.011) eta 1:09:15 loss 1.4392 (1.3461) acc 96.8750 (96.8750) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.4384 (1.3695) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 822.054, TIME@all 0.311 -epoch: [85/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:09:15 loss 1.4547 (1.3645) acc 93.7500 (96.5625) lr 0.260000 -epoch: [85/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:08:56 loss 1.5754 (1.3801) acc 87.5000 (96.0156) lr 0.260000 -FPS@all 822.018, TIME@all 0.311 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.5522 (1.3546) acc 96.8750 (96.2500) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5212 (1.3792) acc 93.7500 (95.2344) lr 0.260000 -FPS@all 821.952, TIME@all 0.311 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:34 loss 1.5045 (1.3224) acc 87.5000 (96.8750) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5095 (1.3527) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 821.987, TIME@all 0.311 -epoch: [86/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.4030 (1.3179) acc 93.7500 (97.6562) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3377 (1.3548) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 821.854, TIME@all 0.311 -epoch: [86/350][20/50] time 0.314 (0.311) data 0.001 (0.012) eta 1:08:35 loss 1.3628 (1.3040) acc 100.0000 (97.1875) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.6072 (1.3486) acc 84.3750 (95.9375) lr 0.260000 -FPS@all 821.883, TIME@all 0.311 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.6154 (1.3585) acc 87.5000 (95.6250) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.5044 (1.3654) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 821.808, TIME@all 0.312 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:35 loss 1.5558 (1.3436) acc 90.6250 (96.5625) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3877 (1.3561) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 821.903, TIME@all 0.311 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 1:08:34 loss 1.6066 (1.3036) acc 90.6250 (97.6562) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.4873 (1.3406) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 821.963, TIME@all 0.311 -epoch: [86/350][20/50] time 0.313 (0.311) data 0.000 (0.013) eta 1:08:35 loss 1.5968 (1.3336) acc 90.6250 (96.7188) lr 0.260000 -epoch: [86/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:38 loss 1.3660 (1.3595) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 821.896, TIME@all 0.311 -epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:08:45 loss 1.2947 (1.3260) acc 96.8750 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3558 (1.3445) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 820.516, TIME@all 0.312 -epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:08:45 loss 1.2934 (1.2975) acc 96.8750 (97.9688) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:08:31 loss 1.2300 (1.3400) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 820.421, TIME@all 0.312 -epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:46 loss 1.2888 (1.2969) acc 96.8750 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2073 (1.3180) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 820.397, TIME@all 0.312 -epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:08:45 loss 1.3411 (1.3156) acc 96.8750 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.4084 (1.3490) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.469, TIME@all 0.312 -epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.3065 (1.3030) acc 90.6250 (96.5625) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3257 (1.3287) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.461, TIME@all 0.312 -epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.3004 (1.3051) acc 96.8750 (97.6562) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.3320 (1.3335) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 820.466, TIME@all 0.312 -epoch: [87/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:08:45 loss 1.2377 (1.2918) acc 100.0000 (97.3438) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2638 (1.3396) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 820.422, TIME@all 0.312 -epoch: [87/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:08:46 loss 1.3157 (1.2886) acc 100.0000 (97.9688) lr 0.260000 -epoch: [87/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:08:31 loss 1.2237 (1.3167) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 820.432, TIME@all 0.312 -epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.3618 (1.3117) acc 93.7500 (97.1875) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.2816 (1.3409) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 821.264, TIME@all 0.312 -epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.3292 (1.2975) acc 96.8750 (97.5000) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.2978 (1.3436) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 821.152, TIME@all 0.312 -epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.2605 (1.3251) acc 100.0000 (96.4062) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.4561 (1.3329) acc 90.6250 (96.4844) lr 0.260000 -FPS@all 821.090, TIME@all 0.312 -epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.4107 (1.3265) acc 96.8750 (96.7188) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.5540 (1.3490) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 821.061, TIME@all 0.312 -epoch: [88/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.4478 (1.3381) acc 93.7500 (96.8750) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.3527 (1.3515) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 821.091, TIME@all 0.312 -epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:20 loss 1.2977 (1.2818) acc 96.8750 (97.9688) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.3046 (1.3125) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.131, TIME@all 0.312 -epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 1:08:19 loss 1.2734 (1.2961) acc 96.8750 (97.6562) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:08:15 loss 1.2902 (1.3348) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 821.115, TIME@all 0.312 -epoch: [88/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 1:08:19 loss 1.3849 (1.3122) acc 93.7500 (97.0312) lr 0.260000 -epoch: [88/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:08:15 loss 1.5208 (1.3422) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 821.128, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.3139 (1.2599) acc 100.0000 (98.5938) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.3787 (1.2714) acc 93.7500 (98.1250) lr 0.260000 -FPS@all 820.261, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2691 (1.2729) acc 100.0000 (97.9688) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:03 loss 1.3913 (1.2903) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 820.270, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.5311 (1.2842) acc 90.6250 (97.6562) lr 0.260000 -epoch: [89/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:08:04 loss 1.6006 (1.2981) acc 84.3750 (97.0312) lr 0.260000 -FPS@all 820.084, TIME@all 0.312 -epoch: [89/350][20/50] time 0.308 (0.313) data 0.000 (0.011) eta 1:08:11 loss 1.1935 (1.2680) acc 96.8750 (98.1250) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4719 (1.2916) acc 87.5000 (97.1094) lr 0.260000 -FPS@all 820.095, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.1824 (1.2609) acc 100.0000 (98.2812) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4329 (1.2853) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 820.179, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2251 (1.2487) acc 100.0000 (99.0625) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4058 (1.2786) acc 90.6250 (97.9688) lr 0.260000 -FPS@all 820.203, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2715 (1.2611) acc 93.7500 (98.9062) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4786 (1.2877) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 820.185, TIME@all 0.312 -epoch: [89/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 1:08:11 loss 1.2685 (1.2568) acc 96.8750 (98.2812) lr 0.260000 -epoch: [89/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 1:08:04 loss 1.4209 (1.2913) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 820.155, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.5439 (1.3559) acc 87.5000 (95.4688) lr 0.260000 -epoch: [90/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 1:07:49 loss 1.3350 (1.3545) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 819.530, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.2299 (1.3141) acc 100.0000 (97.5000) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:49 loss 1.2364 (1.3258) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 819.574, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 1:07:58 loss 1.3307 (1.3318) acc 90.6250 (96.4062) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.2651 (1.3576) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 819.439, TIME@all 0.312 -epoch: [90/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:07:59 loss 1.2100 (1.3257) acc 100.0000 (96.4062) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.1838 (1.3493) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 819.406, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.3190 (1.3040) acc 100.0000 (97.1875) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.2810 (1.3403) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 819.454, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.2129 (1.3055) acc 96.8750 (97.1875) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.3431 (1.3720) acc 96.8750 (95.2344) lr 0.260000 -FPS@all 819.441, TIME@all 0.312 -epoch: [90/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:58 loss 1.4487 (1.3306) acc 96.8750 (97.1875) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:50 loss 1.2261 (1.3624) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 819.422, TIME@all 0.312 -epoch: [90/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:07:59 loss 1.2686 (1.3372) acc 100.0000 (97.5000) lr 0.260000 -epoch: [90/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 1:07:50 loss 1.2824 (1.3513) acc 100.0000 (96.8750) lr 0.260000 -FPS@all 819.457, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:07:39 loss 1.2463 (1.2869) acc 96.8750 (97.6562) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:07:34 loss 1.5287 (1.3194) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 819.914, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3845 (1.2839) acc 96.8750 (98.4375) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:33 loss 1.3609 (1.2940) acc 93.7500 (97.8125) lr 0.260000 -FPS@all 819.982, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:40 loss 1.3602 (1.2967) acc 96.8750 (97.3438) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.4910 (1.3273) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 819.833, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.5551 (1.2931) acc 90.6250 (97.3438) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.6802 (1.3166) acc 84.3750 (96.5625) lr 0.260000 -FPS@all 819.852, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3266 (1.2676) acc 93.7500 (97.8125) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.5103 (1.3038) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 819.840, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3946 (1.2882) acc 90.6250 (97.3438) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:33 loss 1.4482 (1.3071) acc 90.6250 (97.0312) lr 0.260000 -FPS@all 819.894, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.3420 (1.2828) acc 96.8750 (97.9688) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.3411 (1.3006) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 819.852, TIME@all 0.312 -epoch: [91/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:07:39 loss 1.2250 (1.2694) acc 100.0000 (97.9688) lr 0.260000 -epoch: [91/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:07:34 loss 1.2671 (1.3059) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.897, TIME@all 0.312 -epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:07:30 loss 1.2675 (1.2959) acc 100.0000 (97.5000) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.3614 (1.3201) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 819.431, TIME@all 0.312 -epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.3495 (1.2652) acc 96.8750 (97.9688) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.3245 (1.3015) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 819.358, TIME@all 0.312 -epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.4633 (1.3065) acc 96.8750 (97.0312) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:21 loss 1.3757 (1.3337) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 819.279, TIME@all 0.312 -epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:31 loss 1.2792 (1.2922) acc 100.0000 (97.5000) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:21 loss 1.4889 (1.3449) acc 90.6250 (96.1719) lr 0.260000 -FPS@all 819.288, TIME@all 0.312 -epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:31 loss 1.3702 (1.2974) acc 96.8750 (97.5000) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:07:20 loss 1.4746 (1.3393) acc 87.5000 (96.4844) lr 0.260000 -FPS@all 819.321, TIME@all 0.312 -epoch: [92/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.4851 (1.3167) acc 90.6250 (96.5625) lr 0.260000 -epoch: [92/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:07:20 loss 1.4684 (1.3404) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 819.304, TIME@all 0.312 -epoch: [92/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:07:30 loss 1.2580 (1.2920) acc 96.8750 (97.6562) lr 0.260000 -epoch: [92/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:07:20 loss 1.3388 (1.3243) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 819.344, TIME@all 0.312 -epoch: [92/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:07:30 loss 1.3256 (1.2726) acc 96.8750 (98.1250) lr 0.260000 -epoch: [92/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:07:20 loss 1.3903 (1.3098) acc 90.6250 (97.5000) lr 0.260000 -FPS@all 819.281, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3727 (1.3414) acc 96.8750 (96.2500) lr 0.260000 -epoch: [93/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.3546 (1.3767) acc 96.8750 (95.4688) lr 0.260000 -FPS@all 821.150, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3041 (1.3461) acc 100.0000 (97.1875) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.5583 (1.3787) acc 90.6250 (95.9375) lr 0.260000 -FPS@all 821.188, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.4923 (1.3568) acc 93.7500 (96.2500) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.2709 (1.3865) acc 96.8750 (95.1562) lr 0.260000 -FPS@all 821.045, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:07:03 loss 1.3783 (1.3513) acc 93.7500 (96.8750) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.3711 (1.3628) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 821.100, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.4107 (1.3638) acc 93.7500 (95.3125) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.2992 (1.3619) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 821.099, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.2889 (1.3417) acc 100.0000 (96.8750) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.4701 (1.3741) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 821.041, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:07:03 loss 1.3569 (1.3473) acc 93.7500 (96.7188) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.006) eta 1:07:00 loss 1.4054 (1.3564) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.090, TIME@all 0.312 -epoch: [93/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:07:04 loss 1.4200 (1.3471) acc 93.7500 (95.9375) lr 0.260000 -epoch: [93/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 1:07:00 loss 1.2905 (1.3633) acc 100.0000 (96.0156) lr 0.260000 -FPS@all 821.083, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:55 loss 1.2759 (1.2900) acc 96.8750 (97.6562) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:06:42 loss 1.2461 (1.3073) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 820.671, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 1:06:56 loss 1.2898 (1.3100) acc 100.0000 (97.3438) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2879 (1.3147) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 820.498, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:06:56 loss 1.3413 (1.2822) acc 93.7500 (97.0312) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2174 (1.3105) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.501, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3123 (1.2752) acc 96.8750 (97.9688) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:06:42 loss 1.4214 (1.3147) acc 90.6250 (96.7188) lr 0.260000 -FPS@all 820.562, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 1:06:56 loss 1.2816 (1.2916) acc 96.8750 (97.5000) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:43 loss 1.2117 (1.3176) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 820.531, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3561 (1.3048) acc 90.6250 (97.3438) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 1:06:43 loss 1.3406 (1.3221) acc 90.6250 (96.2500) lr 0.260000 -FPS@all 820.493, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.2684 (1.2984) acc 100.0000 (97.8125) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:43 loss 1.3770 (1.3088) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 820.503, TIME@all 0.312 -epoch: [94/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:06:56 loss 1.3013 (1.3001) acc 96.8750 (97.9688) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -epoch: [94/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 1:06:43 loss 1.3277 (1.3074) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 820.525, TIME@all 0.312 -epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.4244 (1.2732) acc 93.7500 (98.5938) lr 0.260000 -epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3628 (1.3148) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 820.571, TIME@all 0.312 -epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2472 (1.2752) acc 100.0000 (98.2812) lr 0.260000 -epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:25 loss 1.3724 (1.3053) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 820.637, TIME@all 0.312 -epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:06:38 loss 1.2208 (1.3050) acc 100.0000 (97.3438) lr 0.260000 -epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.4278 (1.3236) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 820.477, TIME@all 0.312 -epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2183 (1.2467) acc 100.0000 (98.7500) lr 0.260000 -epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3290 (1.2837) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 820.514, TIME@all 0.312 -epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2717 (1.2882) acc 100.0000 (97.3438) lr 0.260000 -epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3624 (1.3248) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 820.497, TIME@all 0.312 -epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:06:39 loss 1.4073 (1.3199) acc 96.8750 (96.4062) lr 0.260000 -epoch: [95/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:06:26 loss 1.3820 (1.3267) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 820.545, TIME@all 0.312 -epoch: [95/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:06:38 loss 1.2084 (1.2715) acc 100.0000 (97.9688) lr 0.260000 -epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3962 (1.3004) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 820.566, TIME@all 0.312 -epoch: [95/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:06:39 loss 1.2627 (1.2680) acc 96.8750 (98.2812) lr 0.260000 -epoch: [95/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:06:26 loss 1.3227 (1.3051) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 820.531, TIME@all 0.312 -epoch: [96/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:06:21 loss 1.4948 (1.3028) acc 93.7500 (97.3438) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.3048 (1.3058) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 820.658, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:06:21 loss 1.3000 (1.3014) acc 96.8750 (97.1875) lr 0.260000 -epoch: [96/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:12 loss 1.2254 (1.2923) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 820.689, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:06:22 loss 1.4184 (1.3096) acc 96.8750 (96.8750) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:13 loss 1.2510 (1.3018) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 820.539, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:06:22 loss 1.3701 (1.2783) acc 96.8750 (97.8125) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:13 loss 1.3490 (1.3048) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 820.593, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:06:22 loss 1.2958 (1.2803) acc 100.0000 (97.3438) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.2604 (1.3267) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 820.570, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.012) eta 1:06:21 loss 1.2972 (1.2665) acc 100.0000 (97.9688) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:06:12 loss 1.2572 (1.2975) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 820.655, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:06:22 loss 1.5020 (1.3043) acc 90.6250 (97.1875) lr 0.260000 -epoch: [96/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:06:13 loss 1.3188 (1.3191) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 820.579, TIME@all 0.312 -epoch: [96/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:06:22 loss 1.3324 (1.3056) acc 96.8750 (98.1250) lr 0.260000 -epoch: [96/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 1:06:13 loss 1.1904 (1.3112) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 820.579, TIME@all 0.312 -epoch: [97/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 1:06:13 loss 1.2285 (1.2724) acc 100.0000 (97.6562) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.3596 (1.3012) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 820.097, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 1:06:13 loss 1.3221 (1.2738) acc 93.7500 (97.6562) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.2931 (1.2966) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 820.158, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:06:13 loss 1.1636 (1.2793) acc 100.0000 (98.4375) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.4218 (1.3095) acc 96.8750 (97.8125) lr 0.260000 -FPS@all 820.031, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 1:06:14 loss 1.2424 (1.2920) acc 100.0000 (97.1875) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.3469 (1.3185) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 820.019, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.012) eta 1:06:13 loss 1.2069 (1.2996) acc 100.0000 (97.1875) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:06:00 loss 1.3672 (1.2982) acc 90.6250 (97.3438) lr 0.260000 -FPS@all 820.054, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 1:06:14 loss 1.2756 (1.2902) acc 93.7500 (97.8125) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:06:00 loss 1.3533 (1.3182) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.058, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:06:13 loss 1.2352 (1.2813) acc 100.0000 (97.9688) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:06:00 loss 1.3637 (1.3092) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 820.046, TIME@all 0.312 -epoch: [97/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 1:06:13 loss 1.2031 (1.2730) acc 100.0000 (97.5000) lr 0.260000 -epoch: [97/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:59 loss 1.3744 (1.3250) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 820.114, TIME@all 0.312 -epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:45 loss 1.5092 (1.3662) acc 90.6250 (94.8438) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:39 loss 1.4616 (1.3971) acc 90.6250 (94.9219) lr 0.260000 -FPS@all 820.462, TIME@all 0.312 -epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.9022 (1.3556) acc 81.2500 (96.4062) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:39 loss 1.2992 (1.3807) acc 96.8750 (95.5469) lr 0.260000 -FPS@all 820.394, TIME@all 0.312 -epoch: [98/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.7564 (1.3665) acc 87.5000 (96.2500) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.2598 (1.3858) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 820.360, TIME@all 0.312 -epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.7568 (1.3720) acc 84.3750 (96.2500) lr 0.260000 -epoch: [98/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4500 (1.4096) acc 90.6250 (95.0781) lr 0.260000 -FPS@all 820.323, TIME@all 0.312 -epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.8032 (1.3666) acc 84.3750 (95.6250) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4621 (1.3699) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 820.373, TIME@all 0.312 -epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.6231 (1.3486) acc 93.7500 (96.2500) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4843 (1.3775) acc 96.8750 (95.7031) lr 0.260000 -FPS@all 820.369, TIME@all 0.312 -epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:05:46 loss 1.6145 (1.3608) acc 87.5000 (96.5625) lr 0.260000 -epoch: [98/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 1:05:40 loss 1.4270 (1.3902) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 820.333, TIME@all 0.312 -epoch: [98/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 1:05:46 loss 1.5668 (1.3669) acc 93.7500 (95.4688) lr 0.260000 -epoch: [98/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 1:05:40 loss 1.4603 (1.3980) acc 96.8750 (95.0781) lr 0.260000 -FPS@all 820.358, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.4851 (1.3300) acc 93.7500 (97.1875) lr 0.260000 -epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.5543 (1.3491) acc 87.5000 (96.4062) lr 0.260000 -FPS@all 820.243, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.312) data 0.000 (0.014) eta 1:05:30 loss 1.2120 (1.3059) acc 100.0000 (97.9688) lr 0.260000 -epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:27 loss 1.2284 (1.3362) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 820.296, TIME@all 0.312 -epoch: [99/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.3125 (1.3045) acc 96.8750 (97.0312) lr 0.260000 -epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.4017 (1.3444) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 820.156, TIME@all 0.312 -epoch: [99/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 1:05:32 loss 1.6102 (1.3425) acc 93.7500 (97.1875) lr 0.260000 -epoch: [99/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:05:28 loss 1.2319 (1.3507) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.139, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.4480 (1.3019) acc 96.8750 (97.3438) lr 0.260000 -epoch: [99/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.4224 (1.3348) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 820.185, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.3979 (1.3247) acc 90.6250 (96.4062) lr 0.260000 -epoch: [99/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.2784 (1.3321) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 820.207, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.2994 (1.2960) acc 96.8750 (97.6562) lr 0.260000 -epoch: [99/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:05:28 loss 1.3217 (1.3244) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 820.199, TIME@all 0.312 -epoch: [99/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:05:31 loss 1.6389 (1.3304) acc 87.5000 (97.1875) lr 0.260000 -epoch: [99/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 1:05:28 loss 1.4749 (1.3413) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 820.168, TIME@all 0.312 -epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:05:08 loss 1.1902 (1.2903) acc 100.0000 (97.1875) lr 0.260000 -epoch: [100/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.3324 (1.3145) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 820.740, TIME@all 0.312 -epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.4519 (1.2964) acc 93.7500 (98.4375) lr 0.260000 -epoch: [100/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.4419 (1.3058) acc 90.6250 (97.7344) lr 0.260000 -FPS@all 820.764, TIME@all 0.312 -epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.5024 (1.2906) acc 90.6250 (97.6562) lr 0.260000 -epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.4901 (1.3308) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.643, TIME@all 0.312 -epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.2125 (1.2887) acc 100.0000 (98.2812) lr 0.260000 -epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2789 (1.3143) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 820.668, TIME@all 0.312 -epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.3967 (1.2757) acc 96.8750 (98.4375) lr 0.260000 -epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2354 (1.3078) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 820.690, TIME@all 0.312 -epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:07 loss 1.3629 (1.2721) acc 93.7500 (98.2812) lr 0.260000 -epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:04 loss 1.3837 (1.3135) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 820.697, TIME@all 0.312 -epoch: [100/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:05:08 loss 1.4032 (1.2787) acc 96.8750 (97.0312) lr 0.260000 -epoch: [100/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:05:05 loss 1.2085 (1.2976) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 820.658, TIME@all 0.312 -epoch: [100/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:05:08 loss 1.3334 (1.3248) acc 96.8750 (96.8750) lr 0.260000 -epoch: [100/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:05:04 loss 1.4364 (1.3361) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 820.698, TIME@all 0.312 -epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.3393 (1.3466) acc 93.7500 (96.5625) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:04 loss 1.3731 (1.3439) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 818.624, TIME@all 0.313 -epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.3456 (1.3022) acc 100.0000 (97.3438) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.2948 (1.3391) acc 100.0000 (96.3281) lr 0.260000 -FPS@all 818.429, TIME@all 0.313 -epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.3008 (1.3352) acc 96.8750 (97.0312) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.3647 (1.3284) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 818.444, TIME@all 0.313 -epoch: [101/350][20/50] time 0.312 (0.314) data 0.001 (0.012) eta 1:05:16 loss 1.3193 (1.3150) acc 96.8750 (95.9375) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:05:05 loss 1.2225 (1.3476) acc 100.0000 (95.7031) lr 0.260000 -FPS@all 818.503, TIME@all 0.313 -epoch: [101/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.2879 (1.2900) acc 93.7500 (97.3438) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.3732 (1.3330) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 818.481, TIME@all 0.313 -epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.2911 (1.2951) acc 96.8750 (97.5000) lr 0.260000 -epoch: [101/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 1:05:04 loss 1.2922 (1.3249) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 818.489, TIME@all 0.313 -epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:17 loss 1.2440 (1.3124) acc 93.7500 (96.8750) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 1:05:05 loss 1.2210 (1.3456) acc 100.0000 (95.8594) lr 0.260000 -FPS@all 818.513, TIME@all 0.313 -epoch: [101/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:05:16 loss 1.3267 (1.3230) acc 96.8750 (96.7188) lr 0.260000 -epoch: [101/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 1:05:04 loss 1.2838 (1.3390) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 818.508, TIME@all 0.313 -epoch: [102/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 1:04:41 loss 1.2926 (1.3163) acc 96.8750 (96.7188) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:32 loss 1.3628 (1.3255) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 821.052, TIME@all 0.312 -epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.4345 (1.3241) acc 93.7500 (97.0312) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3502 (1.3376) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.958, TIME@all 0.312 -epoch: [102/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.3269 (1.3186) acc 100.0000 (96.7188) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3369 (1.3295) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.893, TIME@all 0.312 -epoch: [102/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 1:04:42 loss 1.3068 (1.3129) acc 93.7500 (96.7188) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.3036 (1.3234) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 820.931, TIME@all 0.312 -epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.2831 (1.3192) acc 100.0000 (97.1875) lr 0.260000 -epoch: [102/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.4169 (1.3414) acc 90.6250 (96.7188) lr 0.260000 -FPS@all 820.881, TIME@all 0.312 -epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.2868 (1.3204) acc 100.0000 (97.6562) lr 0.260000 -epoch: [102/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2994 (1.3464) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 820.936, TIME@all 0.312 -epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:42 loss 1.4019 (1.3141) acc 100.0000 (97.1875) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2717 (1.3354) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 820.920, TIME@all 0.312 -epoch: [102/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:04:41 loss 1.2799 (1.3118) acc 96.8750 (97.5000) lr 0.260000 -epoch: [102/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:04:33 loss 1.2555 (1.3354) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 820.976, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:30 loss 1.4086 (1.2671) acc 96.8750 (98.7500) lr 0.260000 -epoch: [103/350][40/50] time 0.321 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3189 (1.3052) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 819.878, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:31 loss 1.2087 (1.2627) acc 100.0000 (97.8125) lr 0.260000 -epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 1:04:25 loss 1.2752 (1.3008) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.884, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.2984 (1.2669) acc 100.0000 (98.9062) lr 0.260000 -epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3200 (1.2981) acc 93.7500 (97.8125) lr 0.260000 -FPS@all 819.793, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.3683 (1.2583) acc 96.8750 (98.4375) lr 0.260000 -epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.2412 (1.2857) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 819.732, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.4226 (1.2895) acc 96.8750 (97.5000) lr 0.260000 -epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.3196 (1.2993) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 819.760, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 1:04:31 loss 1.3887 (1.2705) acc 90.6250 (97.5000) lr 0.260000 -epoch: [103/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 1:04:26 loss 1.2909 (1.2855) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 819.805, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:30 loss 1.2537 (1.2907) acc 96.8750 (97.9688) lr 0.260000 -epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:04:26 loss 1.2879 (1.3019) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 819.835, TIME@all 0.312 -epoch: [103/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:04:31 loss 1.2229 (1.2589) acc 100.0000 (98.2812) lr 0.260000 -epoch: [103/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 1:04:26 loss 1.1632 (1.2827) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 819.816, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.001 (0.014) eta 1:04:20 loss 1.2719 (1.2964) acc 100.0000 (98.2812) lr 0.260000 -epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.3325 (1.3187) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 820.117, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.3341 (1.2934) acc 96.8750 (97.3438) lr 0.260000 -epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.5492 (1.3262) acc 84.3750 (96.3281) lr 0.260000 -FPS@all 819.956, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.2478 (1.3276) acc 100.0000 (97.0312) lr 0.260000 -epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2313 (1.3327) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 819.998, TIME@all 0.312 -epoch: [104/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:04:20 loss 1.4380 (1.2953) acc 93.7500 (96.8750) lr 0.260000 -epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2349 (1.3197) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 820.076, TIME@all 0.312 -epoch: [104/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 1:04:21 loss 1.3406 (1.2760) acc 93.7500 (97.9688) lr 0.260000 -epoch: [104/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:04:10 loss 1.3314 (1.3114) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.964, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:04:20 loss 1.2373 (1.2757) acc 100.0000 (97.9688) lr 0.260000 -epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:04:10 loss 1.3186 (1.3062) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 820.016, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:21 loss 1.3294 (1.2722) acc 96.8750 (97.0312) lr 0.260000 -epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.3229 (1.2967) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 819.966, TIME@all 0.312 -epoch: [104/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:04:20 loss 1.3880 (1.2933) acc 96.8750 (97.3438) lr 0.260000 -epoch: [104/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:04:10 loss 1.2493 (1.3163) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 820.013, TIME@all 0.312 -epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:51 loss 1.2672 (1.2763) acc 96.8750 (97.9688) lr 0.260000 -epoch: [105/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.2732 (1.2900) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 821.076, TIME@all 0.312 -epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:51 loss 1.3242 (1.2554) acc 96.8750 (97.6562) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.2837 (1.2940) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.141, TIME@all 0.312 -epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.4335 (1.2972) acc 96.8750 (97.5000) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.3193 (1.3119) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 821.000, TIME@all 0.312 -epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.3881 (1.2926) acc 90.6250 (97.0312) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:48 loss 1.3080 (1.2947) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 820.974, TIME@all 0.312 -epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:03:52 loss 1.4120 (1.2869) acc 96.8750 (97.9688) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:48 loss 1.3797 (1.3168) acc 84.3750 (96.6406) lr 0.260000 -FPS@all 821.028, TIME@all 0.312 -epoch: [105/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:03:51 loss 1.1946 (1.2665) acc 100.0000 (98.2812) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.4341 (1.3031) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 820.973, TIME@all 0.312 -epoch: [105/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:03:52 loss 1.3665 (1.3069) acc 96.8750 (96.8750) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:03:48 loss 1.2790 (1.3295) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 821.000, TIME@all 0.312 -epoch: [105/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 1:03:51 loss 1.2862 (1.2930) acc 96.8750 (97.5000) lr 0.260000 -epoch: [105/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 1:03:47 loss 1.3939 (1.3216) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 821.084, TIME@all 0.312 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.001 (0.012) eta 1:04:16 loss 1.2748 (1.3129) acc 100.0000 (97.8125) lr 0.260000 -epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:07 loss 1.2851 (1.3007) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 815.525, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 1:04:16 loss 1.3774 (1.3139) acc 93.7500 (97.3438) lr 0.260000 -epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.007) eta 1:04:07 loss 1.3480 (1.2978) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 815.531, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:17 loss 1.3921 (1.2808) acc 96.8750 (98.4375) lr 0.260000 -epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.2208 (1.3073) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 815.369, TIME@all 0.314 -epoch: [106/350][20/50] time 0.313 (0.315) data 0.001 (0.012) eta 1:04:16 loss 1.2403 (1.2562) acc 100.0000 (99.2188) lr 0.260000 -epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.2939 (1.2915) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 815.474, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:17 loss 1.3334 (1.2879) acc 100.0000 (98.2812) lr 0.260000 -epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.3187 (1.3002) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 815.376, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 1:04:16 loss 1.2896 (1.2752) acc 100.0000 (98.7500) lr 0.260000 -epoch: [106/350][40/50] time 0.324 (0.315) data 0.000 (0.006) eta 1:04:08 loss 1.3388 (1.3120) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 815.433, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 1:04:16 loss 1.3297 (1.2963) acc 96.8750 (97.6562) lr 0.260000 -epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 1:04:08 loss 1.3065 (1.3038) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 815.446, TIME@all 0.314 -epoch: [106/350][20/50] time 0.312 (0.315) data 0.001 (0.013) eta 1:04:14 loss 1.3725 (1.3231) acc 96.8750 (96.4062) lr 0.260000 -epoch: [106/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 1:04:07 loss 1.4628 (1.3243) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 815.607, TIME@all 0.314 -epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:21 loss 1.2989 (1.3096) acc 96.8750 (97.1875) lr 0.260000 -epoch: [107/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3523 (1.3125) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 821.353, TIME@all 0.312 -epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.2416 (1.2991) acc 100.0000 (97.3438) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:17 loss 1.4005 (1.3343) acc 90.6250 (96.1719) lr 0.260000 -FPS@all 821.200, TIME@all 0.312 -epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.2605 (1.2976) acc 96.8750 (96.7188) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3084 (1.3076) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 821.256, TIME@all 0.312 -epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.011) eta 1:03:22 loss 1.2385 (1.2858) acc 100.0000 (97.9688) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3749 (1.3060) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 821.182, TIME@all 0.312 -epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.3015 (1.2731) acc 96.8750 (97.1875) lr 0.260000 -epoch: [107/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3953 (1.2955) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 821.249, TIME@all 0.312 -epoch: [107/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:03:21 loss 1.3987 (1.2931) acc 90.6250 (97.1875) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.3121 (1.2947) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 821.259, TIME@all 0.312 -epoch: [107/350][20/50] time 0.309 (0.312) data 0.001 (0.012) eta 1:03:22 loss 1.3501 (1.2940) acc 96.8750 (97.8125) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 1:03:16 loss 1.2283 (1.3140) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 821.242, TIME@all 0.312 -epoch: [107/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 1:03:22 loss 1.1985 (1.2672) acc 100.0000 (97.9688) lr 0.260000 -epoch: [107/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 1:03:16 loss 1.3753 (1.2925) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 821.261, TIME@all 0.312 -epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:03:06 loss 1.2928 (1.2514) acc 100.0000 (98.2812) lr 0.260000 -epoch: [108/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3675 (1.2998) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 821.598, TIME@all 0.312 -epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 1:03:05 loss 1.3215 (1.2817) acc 100.0000 (98.1250) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3468 (1.3032) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 821.666, TIME@all 0.312 -epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:03:06 loss 1.2597 (1.2765) acc 96.8750 (97.6562) lr 0.260000 -epoch: [108/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 1:02:58 loss 1.4482 (1.3111) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 821.486, TIME@all 0.312 -epoch: [108/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 1:03:05 loss 1.3995 (1.2756) acc 93.7500 (97.8125) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3653 (1.2983) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 821.547, TIME@all 0.312 -epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:03:05 loss 1.3464 (1.2906) acc 96.8750 (97.1875) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3258 (1.3197) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 821.613, TIME@all 0.312 -epoch: [108/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 1:03:05 loss 1.2986 (1.2982) acc 100.0000 (97.9688) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 1:02:57 loss 1.3033 (1.3328) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 821.596, TIME@all 0.312 -epoch: [108/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:03:06 loss 1.2895 (1.2709) acc 96.8750 (97.9688) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3372 (1.3036) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 821.590, TIME@all 0.312 -epoch: [108/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 1:03:06 loss 1.3147 (1.2857) acc 100.0000 (97.0312) lr 0.260000 -epoch: [108/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 1:02:57 loss 1.3845 (1.3232) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 821.570, TIME@all 0.312 -epoch: [109/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.4079 (1.2672) acc 87.5000 (97.1875) lr 0.260000 -epoch: [109/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2820 (1.2770) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 819.372, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.2075 (1.2623) acc 100.0000 (97.5000) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2956 (1.2778) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 819.407, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:57 loss 1.2983 (1.2730) acc 100.0000 (97.9688) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.2543 (1.2759) acc 100.0000 (98.4375) lr 0.260000 -FPS@all 819.244, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 1:02:57 loss 1.2580 (1.2536) acc 100.0000 (98.1250) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 1:02:54 loss 1.2125 (1.2843) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 819.252, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.2619 (1.2539) acc 100.0000 (98.2812) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.2680 (1.2628) acc 93.7500 (98.0469) lr 0.260000 -FPS@all 819.350, TIME@all 0.312 -epoch: [109/350][20/50] time 0.315 (0.313) data 0.001 (0.013) eta 1:02:56 loss 1.2335 (1.2470) acc 100.0000 (98.1250) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:53 loss 1.3225 (1.2693) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 819.297, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 1:02:56 loss 1.3270 (1.2614) acc 96.8750 (98.1250) lr 0.260000 -epoch: [109/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.2867 (1.2681) acc 100.0000 (98.0469) lr 0.260000 -FPS@all 819.329, TIME@all 0.312 -epoch: [109/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 1:02:56 loss 1.2710 (1.2659) acc 96.8750 (97.6562) lr 0.260000 -epoch: [109/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:02:54 loss 1.3296 (1.2674) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 819.328, TIME@all 0.312 -epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2390 (1.2771) acc 100.0000 (97.9688) lr 0.260000 -epoch: [110/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:02:25 loss 1.2898 (1.3170) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 821.918, TIME@all 0.311 -epoch: [110/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:02:35 loss 1.4564 (1.2718) acc 93.7500 (97.6562) lr 0.260000 -epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2603 (1.3107) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 821.817, TIME@all 0.312 -epoch: [110/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.1896 (1.2812) acc 100.0000 (97.8125) lr 0.260000 -epoch: [110/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2469 (1.3128) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 821.728, TIME@all 0.312 -epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 1:02:34 loss 1.2467 (1.2515) acc 96.8750 (98.4375) lr 0.260000 -epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 1:02:26 loss 1.3323 (1.2910) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 821.771, TIME@all 0.312 -epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.4556 (1.2660) acc 90.6250 (98.1250) lr 0.260000 -epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.3416 (1.3183) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 821.736, TIME@all 0.312 -epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2264 (1.2929) acc 100.0000 (97.6562) lr 0.260000 -epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.2662 (1.3195) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 821.802, TIME@all 0.312 -epoch: [110/350][20/50] time 0.306 (0.312) data 0.000 (0.013) eta 1:02:34 loss 1.6545 (1.2957) acc 87.5000 (97.3438) lr 0.260000 -epoch: [110/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 1:02:26 loss 1.2786 (1.3213) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 821.779, TIME@all 0.312 -epoch: [110/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 1:02:34 loss 1.2798 (1.2516) acc 96.8750 (98.5938) lr 0.260000 -epoch: [110/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 1:02:26 loss 1.3159 (1.2933) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 821.740, TIME@all 0.312 -epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:02:20 loss 1.2775 (1.2605) acc 100.0000 (98.2812) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.3773 (1.2753) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 819.336, TIME@all 0.312 -epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2542 (1.2555) acc 100.0000 (98.5938) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2964 (1.2806) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 819.318, TIME@all 0.312 -epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2380 (1.2762) acc 100.0000 (98.2812) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2824 (1.3002) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 819.312, TIME@all 0.312 -epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2951 (1.2475) acc 100.0000 (98.7500) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2150 (1.2700) acc 100.0000 (98.0469) lr 0.260000 -FPS@all 819.378, TIME@all 0.312 -epoch: [111/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2037 (1.2354) acc 100.0000 (98.7500) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.4882 (1.2713) acc 90.6250 (97.7344) lr 0.260000 -FPS@all 819.336, TIME@all 0.312 -epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:21 loss 1.3993 (1.2438) acc 93.7500 (98.7500) lr 0.260000 -epoch: [111/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.2614 (1.2814) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 819.296, TIME@all 0.312 -epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 1:02:20 loss 1.1800 (1.2370) acc 100.0000 (99.0625) lr 0.260000 -epoch: [111/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:02:25 loss 1.2677 (1.2843) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 819.353, TIME@all 0.312 -epoch: [111/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 1:02:20 loss 1.2120 (1.2422) acc 100.0000 (98.9062) lr 0.260000 -epoch: [111/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 1:02:25 loss 1.3454 (1.2790) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 819.310, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.3796 (1.2683) acc 96.8750 (97.9688) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:58 loss 1.2161 (1.3164) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 821.434, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 1:02:07 loss 1.1972 (1.2471) acc 100.0000 (98.7500) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.1925 (1.2759) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 821.379, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 1:02:07 loss 1.3615 (1.2810) acc 93.7500 (96.5625) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 1:01:59 loss 1.2362 (1.3029) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 821.296, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:02:08 loss 1.3708 (1.2775) acc 100.0000 (97.3438) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2891 (1.2968) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 821.244, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.3382 (1.2364) acc 100.0000 (98.5938) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2246 (1.2744) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 821.330, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.2269 (1.2665) acc 100.0000 (98.5938) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2282 (1.2804) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 821.319, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:02:08 loss 1.3315 (1.2616) acc 100.0000 (98.5938) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.1708 (1.2857) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 821.250, TIME@all 0.312 -epoch: [112/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 1:02:07 loss 1.2634 (1.2681) acc 96.8750 (97.5000) lr 0.260000 -epoch: [112/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 1:01:59 loss 1.2294 (1.2981) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 821.326, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 1:01:52 loss 1.3244 (1.3045) acc 100.0000 (97.8125) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2792 (1.3121) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 820.674, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:01:52 loss 1.2143 (1.2646) acc 100.0000 (98.4375) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2731 (1.2920) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 820.589, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:52 loss 1.2775 (1.2975) acc 100.0000 (96.8750) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3135 (1.3067) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 820.544, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:53 loss 1.2692 (1.2856) acc 100.0000 (97.6562) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3105 (1.3112) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 820.457, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:52 loss 1.3065 (1.2824) acc 96.8750 (97.8125) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.2250 (1.2967) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 820.556, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 1:01:53 loss 1.3436 (1.2942) acc 93.7500 (97.5000) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.2478 (1.3075) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 820.514, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 1:01:52 loss 1.3839 (1.2913) acc 90.6250 (96.8750) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:48 loss 1.3872 (1.3134) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 820.540, TIME@all 0.312 -epoch: [113/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 1:01:53 loss 1.2050 (1.2838) acc 100.0000 (97.3438) lr 0.260000 -epoch: [113/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 1:01:49 loss 1.3117 (1.3048) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 820.532, TIME@all 0.312 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:01:52 loss 1.4046 (1.2909) acc 90.6250 (97.6562) lr 0.260000 -epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 1:01:43 loss 1.3622 (1.3257) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 818.185, TIME@all 0.313 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3800 (1.2879) acc 90.6250 (97.5000) lr 0.260000 -epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.3842 (1.3142) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 818.025, TIME@all 0.313 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3237 (1.2854) acc 93.7500 (97.9688) lr 0.260000 -epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.2472 (1.2967) acc 100.0000 (97.8125) lr 0.260000 -FPS@all 817.990, TIME@all 0.313 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.4853 (1.2928) acc 93.7500 (97.5000) lr 0.260000 -epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3297 (1.2980) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 818.080, TIME@all 0.313 -epoch: [114/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3126 (1.2998) acc 96.8750 (98.2812) lr 0.260000 -epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3668 (1.3136) acc 96.8750 (97.8906) lr 0.260000 -FPS@all 818.069, TIME@all 0.313 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 1:01:52 loss 1.4265 (1.2916) acc 96.8750 (97.6562) lr 0.260000 -epoch: [114/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.2445 (1.3211) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 818.061, TIME@all 0.313 -epoch: [114/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.3725 (1.2887) acc 96.8750 (97.1875) lr 0.260000 -epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:44 loss 1.4530 (1.3271) acc 93.7500 (96.4844) lr 0.260000 -FPS@all 818.068, TIME@all 0.313 -epoch: [114/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 1:01:52 loss 1.5391 (1.3278) acc 87.5000 (96.4062) lr 0.260000 -epoch: [114/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 1:01:43 loss 1.3430 (1.3268) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 818.098, TIME@all 0.313 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.2804 (1.2959) acc 100.0000 (97.5000) lr 0.260000 -epoch: [115/350][40/50] time 0.314 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.2491 (1.3378) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 814.499, TIME@all 0.314 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.3182 (1.2841) acc 100.0000 (98.1250) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.007) eta 1:01:39 loss 1.4364 (1.3273) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 814.551, TIME@all 0.314 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:42 loss 1.4814 (1.3285) acc 84.3750 (96.4062) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.4380 (1.3397) acc 93.7500 (96.1719) lr 0.260000 -FPS@all 814.397, TIME@all 0.314 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 1:01:43 loss 1.4958 (1.3269) acc 93.7500 (96.7188) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.4416 (1.3369) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 814.404, TIME@all 0.314 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:43 loss 1.4680 (1.3086) acc 96.8750 (97.1875) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.001 (0.007) eta 1:01:40 loss 1.3314 (1.3367) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 814.434, TIME@all 0.314 -epoch: [115/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:01:42 loss 1.4629 (1.3321) acc 90.6250 (96.5625) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.3963 (1.3730) acc 96.8750 (94.8438) lr 0.260000 -FPS@all 814.504, TIME@all 0.314 -epoch: [115/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 1:01:43 loss 1.3783 (1.2959) acc 96.8750 (97.5000) lr 0.260000 -epoch: [115/350][40/50] time 0.316 (0.315) data 0.000 (0.007) eta 1:01:40 loss 1.3888 (1.3176) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 814.413, TIME@all 0.314 -epoch: [115/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 1:01:42 loss 1.3629 (1.2881) acc 96.8750 (97.9688) lr 0.260000 -epoch: [115/350][40/50] time 0.315 (0.315) data 0.000 (0.006) eta 1:01:40 loss 1.3878 (1.3269) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 814.450, TIME@all 0.314 -epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.2708 (1.3306) acc 100.0000 (97.3438) lr 0.260000 -epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:58 loss 1.3687 (1.3518) acc 100.0000 (96.2500) lr 0.260000 -FPS@all 808.942, TIME@all 0.316 -epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.013) eta 1:01:55 loss 1.2462 (1.3113) acc 100.0000 (96.7188) lr 0.260000 -epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.007) eta 1:01:58 loss 1.3786 (1.3251) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 808.996, TIME@all 0.316 -epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3164 (1.3163) acc 96.8750 (96.8750) lr 0.260000 -epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.006) eta 1:01:58 loss 1.4332 (1.3426) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 808.918, TIME@all 0.316 -epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.012) eta 1:01:56 loss 1.2616 (1.2947) acc 96.8750 (97.0312) lr 0.260000 -epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.4820 (1.3417) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 808.857, TIME@all 0.316 -epoch: [116/350][20/50] time 0.314 (0.317) data 0.000 (0.013) eta 1:01:56 loss 1.3617 (1.3070) acc 96.8750 (96.7188) lr 0.260000 -epoch: [116/350][40/50] time 0.321 (0.318) data 0.001 (0.007) eta 1:01:59 loss 1.3308 (1.3298) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 808.903, TIME@all 0.316 -epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.2199 (1.3005) acc 100.0000 (97.5000) lr 0.260000 -epoch: [116/350][40/50] time 0.321 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.6124 (1.3370) acc 90.6250 (96.4062) lr 0.260000 -FPS@all 808.935, TIME@all 0.316 -epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3787 (1.2980) acc 96.8750 (97.3438) lr 0.260000 -epoch: [116/350][40/50] time 0.321 (0.318) data 0.001 (0.006) eta 1:01:58 loss 1.2107 (1.3307) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 809.061, TIME@all 0.316 -epoch: [116/350][20/50] time 0.313 (0.317) data 0.000 (0.012) eta 1:01:55 loss 1.3660 (1.3456) acc 96.8750 (97.0312) lr 0.260000 -epoch: [116/350][40/50] time 0.322 (0.318) data 0.000 (0.006) eta 1:01:59 loss 1.2776 (1.3398) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 808.928, TIME@all 0.316 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.4747 (1.3426) acc 93.7500 (97.0312) lr 0.260000 -epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:49 loss 1.3984 (1.3744) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 820.078, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:00:52 loss 1.3679 (1.3169) acc 93.7500 (97.0312) lr 0.260000 -epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:50 loss 1.3606 (1.3570) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 819.999, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.6747 (1.3392) acc 84.3750 (96.8750) lr 0.260000 -epoch: [117/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.3275 (1.3514) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 820.035, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5852 (1.3146) acc 90.6250 (97.1875) lr 0.260000 -epoch: [117/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.2231 (1.3554) acc 100.0000 (95.7812) lr 0.260000 -FPS@all 819.987, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5490 (1.3395) acc 93.7500 (96.0938) lr 0.260000 -epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.3245 (1.3590) acc 100.0000 (95.2344) lr 0.260000 -FPS@all 820.040, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.5028 (1.3347) acc 93.7500 (97.6562) lr 0.260000 -epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.4873 (1.3742) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 820.033, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 1:00:52 loss 1.4668 (1.3120) acc 93.7500 (97.0312) lr 0.260000 -epoch: [117/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:50 loss 1.2063 (1.3408) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 820.061, TIME@all 0.312 -epoch: [117/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 1:00:52 loss 1.4449 (1.3136) acc 93.7500 (97.3438) lr 0.260000 -epoch: [117/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 1:00:50 loss 1.2417 (1.3361) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 820.032, TIME@all 0.312 -epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.2721 (1.3580) acc 100.0000 (95.6250) lr 0.260000 -epoch: [118/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3493 (1.3868) acc 93.7500 (95.0781) lr 0.260000 -FPS@all 819.375, TIME@all 0.312 -epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.011) eta 1:00:46 loss 1.4088 (1.3676) acc 93.7500 (95.9375) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.2988 (1.3649) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 819.250, TIME@all 0.312 -epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3231 (1.3699) acc 93.7500 (95.6250) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3168 (1.3950) acc 100.0000 (95.6250) lr 0.260000 -FPS@all 819.318, TIME@all 0.312 -epoch: [118/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3379 (1.3293) acc 96.8750 (97.9688) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3785 (1.3671) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 819.215, TIME@all 0.312 -epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:46 loss 1.3007 (1.3496) acc 96.8750 (96.2500) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.3858 (1.3662) acc 96.8750 (95.7812) lr 0.260000 -FPS@all 819.299, TIME@all 0.312 -epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:47 loss 1.5250 (1.3488) acc 93.7500 (96.7188) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 1:00:37 loss 1.3251 (1.3813) acc 100.0000 (95.4688) lr 0.260000 -FPS@all 819.252, TIME@all 0.312 -epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 1:00:47 loss 1.3959 (1.3178) acc 96.8750 (96.7188) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 1:00:37 loss 1.4385 (1.3757) acc 93.7500 (94.8438) lr 0.260000 -FPS@all 819.240, TIME@all 0.312 -epoch: [118/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 1:00:46 loss 1.2193 (1.3481) acc 100.0000 (96.7188) lr 0.260000 -epoch: [118/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 1:00:37 loss 1.4089 (1.3719) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 819.266, TIME@all 0.312 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3287 (1.2880) acc 100.0000 (97.3438) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3692 (1.3123) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 819.292, TIME@all 0.312 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.012) eta 1:00:25 loss 1.2496 (1.3192) acc 96.8750 (96.5625) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 1:00:20 loss 1.2500 (1.3276) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 819.170, TIME@all 0.313 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3323 (1.2889) acc 93.7500 (97.3438) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.2911 (1.3215) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.173, TIME@all 0.313 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.2662 (1.2880) acc 100.0000 (97.5000) lr 0.260000 -epoch: [119/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3449 (1.3266) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 819.240, TIME@all 0.312 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.1858 (1.2819) acc 100.0000 (97.1875) lr 0.260000 -epoch: [119/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3603 (1.3055) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 819.188, TIME@all 0.313 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.2217 (1.3051) acc 100.0000 (97.6562) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.1926 (1.3307) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 819.196, TIME@all 0.313 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 1:00:25 loss 1.3093 (1.3136) acc 100.0000 (96.7188) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3814 (1.3332) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 819.161, TIME@all 0.313 -epoch: [119/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 1:00:25 loss 1.2437 (1.2838) acc 100.0000 (98.1250) lr 0.260000 -epoch: [119/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 1:00:20 loss 1.3340 (1.3137) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 819.166, TIME@all 0.313 -epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.2227 (1.2840) acc 100.0000 (97.5000) lr 0.260000 -epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:59:49 loss 1.4357 (1.2975) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 822.018, TIME@all 0.311 -epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 1:00:05 loss 1.3144 (1.2915) acc 96.8750 (98.1250) lr 0.260000 -epoch: [120/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:59:48 loss 1.3570 (1.3045) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 822.116, TIME@all 0.311 -epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.2935 (1.2916) acc 96.8750 (97.1875) lr 0.260000 -epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2932 (1.2945) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 821.939, TIME@all 0.311 -epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.2632 (1.2841) acc 96.8750 (97.9688) lr 0.260000 -epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:50 loss 1.3486 (1.3051) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.938, TIME@all 0.311 -epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.2990 (1.2686) acc 100.0000 (98.1250) lr 0.260000 -epoch: [120/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:59:49 loss 1.3681 (1.2904) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 821.992, TIME@all 0.311 -epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 1:00:05 loss 1.2574 (1.2848) acc 96.8750 (97.8125) lr 0.260000 -epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2902 (1.2953) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 822.017, TIME@all 0.311 -epoch: [120/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 1:00:06 loss 1.3383 (1.2667) acc 93.7500 (98.1250) lr 0.260000 -epoch: [120/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:59:49 loss 1.3439 (1.2980) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 821.961, TIME@all 0.311 -epoch: [120/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 1:00:06 loss 1.4026 (1.2822) acc 96.8750 (97.0312) lr 0.260000 -epoch: [120/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:59:49 loss 1.2744 (1.2920) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.979, TIME@all 0.311 -epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2943 (1.2913) acc 100.0000 (97.3438) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2831 (1.3162) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 819.354, TIME@all 0.312 -epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.014) eta 0:59:52 loss 1.3365 (1.2836) acc 96.8750 (97.3438) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:59:45 loss 1.3285 (1.3047) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 819.379, TIME@all 0.312 -epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2240 (1.2517) acc 96.8750 (97.9688) lr 0.260000 -epoch: [121/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2618 (1.2798) acc 100.0000 (97.5781) lr 0.260000 -FPS@all 819.239, TIME@all 0.312 -epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3446 (1.2604) acc 96.8750 (97.8125) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.3106 (1.2866) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 819.287, TIME@all 0.312 -epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3294 (1.2741) acc 96.8750 (97.6562) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:59:46 loss 1.4999 (1.2951) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 819.222, TIME@all 0.312 -epoch: [121/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.2792 (1.2616) acc 96.8750 (97.6562) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:59:46 loss 1.3357 (1.3007) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 819.331, TIME@all 0.312 -epoch: [121/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:59:52 loss 1.3560 (1.2941) acc 100.0000 (97.8125) lr 0.260000 -epoch: [121/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2422 (1.3239) acc 100.0000 (96.5625) lr 0.260000 -FPS@all 819.285, TIME@all 0.312 -epoch: [121/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 0:59:52 loss 1.2416 (1.2523) acc 96.8750 (98.1250) lr 0.260000 -epoch: [121/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:46 loss 1.2670 (1.2792) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 819.318, TIME@all 0.312 -epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:59:48 loss 1.2832 (1.2386) acc 96.8750 (98.7500) lr 0.260000 -epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:59:33 loss 1.5490 (1.2933) acc 90.6250 (97.1875) lr 0.260000 -FPS@all 819.625, TIME@all 0.312 -epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:59:48 loss 1.2892 (1.2539) acc 100.0000 (98.4375) lr 0.260000 -epoch: [122/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:59:32 loss 1.3166 (1.2846) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 819.680, TIME@all 0.312 -epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.3301 (1.2542) acc 93.7500 (97.6562) lr 0.260000 -epoch: [122/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4414 (1.2954) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.484, TIME@all 0.312 -epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:48 loss 1.2465 (1.2565) acc 100.0000 (98.5938) lr 0.260000 -epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4383 (1.2949) acc 93.7500 (97.5000) lr 0.260000 -FPS@all 819.541, TIME@all 0.312 -epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:48 loss 1.3643 (1.2539) acc 96.8750 (98.4375) lr 0.260000 -epoch: [122/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.3212 (1.2803) acc 96.8750 (97.7344) lr 0.260000 -FPS@all 819.560, TIME@all 0.312 -epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:59:49 loss 1.3249 (1.2661) acc 93.7500 (97.9688) lr 0.260000 -epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.3736 (1.2939) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 819.537, TIME@all 0.312 -epoch: [122/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.2685 (1.2629) acc 100.0000 (98.2812) lr 0.260000 -epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.2875 (1.2893) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 819.590, TIME@all 0.312 -epoch: [122/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:59:49 loss 1.2169 (1.2429) acc 100.0000 (98.1250) lr 0.260000 -epoch: [122/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:59:33 loss 1.4188 (1.2671) acc 96.8750 (98.0469) lr 0.260000 -FPS@all 819.563, TIME@all 0.312 -epoch: [123/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.1763 (1.2905) acc 100.0000 (97.1875) lr 0.260000 -epoch: [123/350][40/50] time 0.318 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3944 (1.3205) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 813.685, TIME@all 0.315 -epoch: [123/350][20/50] time 0.311 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.3329 (1.3120) acc 96.8750 (96.7188) lr 0.260000 -epoch: [123/350][40/50] time 0.318 (0.316) data 0.001 (0.007) eta 0:59:45 loss 1.4059 (1.3452) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 813.714, TIME@all 0.315 -epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:59:47 loss 1.2048 (1.3026) acc 100.0000 (97.0312) lr 0.260000 -epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.007) eta 0:59:45 loss 1.4569 (1.3276) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 813.655, TIME@all 0.315 -epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.1936 (1.3006) acc 100.0000 (97.1875) lr 0.260000 -epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3491 (1.3301) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 813.561, TIME@all 0.315 -epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:46 loss 1.2879 (1.3199) acc 100.0000 (96.5625) lr 0.260000 -epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:44 loss 1.4020 (1.3289) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 813.762, TIME@all 0.315 -epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:59:48 loss 1.3763 (1.3129) acc 96.8750 (96.8750) lr 0.260000 -epoch: [123/350][40/50] time 0.318 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.3451 (1.3347) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 813.579, TIME@all 0.315 -epoch: [123/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.2538 (1.2984) acc 100.0000 (97.6562) lr 0.260000 -epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.4514 (1.3290) acc 93.7500 (96.8750) lr 0.260000 -FPS@all 813.620, TIME@all 0.315 -epoch: [123/350][20/50] time 0.313 (0.315) data 0.000 (0.012) eta 0:59:47 loss 1.3274 (1.2936) acc 96.8750 (97.3438) lr 0.260000 -epoch: [123/350][40/50] time 0.317 (0.316) data 0.000 (0.006) eta 0:59:45 loss 1.5061 (1.3250) acc 87.5000 (96.3281) lr 0.260000 -FPS@all 813.625, TIME@all 0.315 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:00 loss 1.6863 (1.3255) acc 84.3750 (96.2500) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3776 (1.3461) acc 96.8750 (95.8594) lr 0.260000 -FPS@all 820.139, TIME@all 0.312 -epoch: [124/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:59:00 loss 1.2946 (1.2852) acc 93.7500 (98.2812) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3675 (1.3167) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 820.195, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.2896 (1.2927) acc 96.8750 (97.8125) lr 0.260000 -epoch: [124/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.2710 (1.3084) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 820.021, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3264 (1.2989) acc 96.8750 (98.4375) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.2523 (1.3132) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 820.088, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.4306 (1.3033) acc 90.6250 (97.3438) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.4281 (1.3061) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 820.018, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3743 (1.2949) acc 96.8750 (97.0312) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3207 (1.3229) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 820.090, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.4536 (1.3073) acc 93.7500 (97.0312) lr 0.260000 -epoch: [124/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:58:56 loss 1.2681 (1.3346) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 820.060, TIME@all 0.312 -epoch: [124/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:59:01 loss 1.3449 (1.3000) acc 96.8750 (97.1875) lr 0.260000 -epoch: [124/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:55 loss 1.3528 (1.3249) acc 96.8750 (96.7188) lr 0.260000 -FPS@all 820.082, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:44 loss 1.4814 (1.3093) acc 93.7500 (97.3438) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.3673 (1.3261) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 820.561, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.3864 (1.3240) acc 96.8750 (96.8750) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.3576 (1.3376) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 820.441, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:44 loss 1.3604 (1.3092) acc 93.7500 (98.2812) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.3359 (1.3231) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 820.612, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.4583 (1.3077) acc 96.8750 (97.3438) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2696 (1.3332) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 820.424, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:58:45 loss 1.5431 (1.3267) acc 93.7500 (97.0312) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.001 (0.007) eta 0:58:38 loss 1.3938 (1.3258) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 820.505, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:58:45 loss 1.5118 (1.3092) acc 96.8750 (97.8125) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:58:38 loss 1.4696 (1.3400) acc 87.5000 (96.6406) lr 0.260000 -FPS@all 820.523, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:58:45 loss 1.4297 (1.3231) acc 96.8750 (96.5625) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2805 (1.3232) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 820.510, TIME@all 0.312 -epoch: [125/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:58:45 loss 1.2553 (1.3056) acc 100.0000 (96.8750) lr 0.260000 -epoch: [125/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:58:38 loss 1.2579 (1.3299) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.490, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.3136 (1.2970) acc 96.8750 (97.8125) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.4177 (1.3058) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 820.366, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:58:26 loss 1.2778 (1.2775) acc 100.0000 (97.9688) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.2453 (1.2821) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 820.435, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:28 loss 1.3298 (1.3002) acc 93.7500 (97.8125) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:58:23 loss 1.3278 (1.3057) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 820.255, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.2207 (1.2841) acc 96.8750 (97.6562) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:22 loss 1.3536 (1.2981) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 820.314, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:28 loss 1.1982 (1.2662) acc 100.0000 (97.9688) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.4208 (1.2860) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 820.281, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:27 loss 1.3801 (1.2865) acc 93.7500 (97.6562) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:58:23 loss 1.4050 (1.3061) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 820.247, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.3163 (1.2882) acc 100.0000 (96.8750) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.2808 (1.3070) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 820.311, TIME@all 0.312 -epoch: [126/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:58:27 loss 1.2743 (1.2930) acc 100.0000 (97.6562) lr 0.260000 -epoch: [126/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:58:23 loss 1.3121 (1.2992) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 820.311, TIME@all 0.312 -epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.3856 (1.2713) acc 96.8750 (97.5000) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2474 (1.3012) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 819.990, TIME@all 0.312 -epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:58:06 loss 1.4688 (1.2812) acc 93.7500 (97.5000) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:58:06 loss 1.2432 (1.2873) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 820.027, TIME@all 0.312 -epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.3163 (1.2685) acc 96.8750 (98.2812) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.4103 (1.3012) acc 93.7500 (97.6562) lr 0.260000 -FPS@all 819.859, TIME@all 0.312 -epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.2864 (1.2756) acc 100.0000 (97.9688) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.2566 (1.2865) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 819.881, TIME@all 0.312 -epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.4809 (1.2827) acc 96.8750 (97.9688) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2263 (1.3057) acc 100.0000 (97.3438) lr 0.260000 -FPS@all 819.965, TIME@all 0.312 -epoch: [127/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.3705 (1.2964) acc 96.8750 (97.6562) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.001 (0.006) eta 0:58:07 loss 1.2245 (1.3093) acc 100.0000 (97.0312) lr 0.260000 -FPS@all 819.918, TIME@all 0.312 -epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:07 loss 1.4127 (1.2741) acc 93.7500 (97.1875) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:06 loss 1.2143 (1.2968) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 819.939, TIME@all 0.312 -epoch: [127/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:58:06 loss 1.2780 (1.2625) acc 96.8750 (98.2812) lr 0.260000 -epoch: [127/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:58:07 loss 1.3674 (1.2962) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 819.928, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.3176 (1.2681) acc 96.8750 (97.9688) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:57:53 loss 1.4592 (1.3228) acc 90.6250 (96.7188) lr 0.260000 -FPS@all 820.873, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.1936 (1.2852) acc 100.0000 (97.6562) lr 0.260000 -epoch: [128/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:57:52 loss 1.6532 (1.3456) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 820.924, TIME@all 0.312 -epoch: [128/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.2375 (1.3140) acc 96.8750 (96.7188) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.3706 (1.3632) acc 93.7500 (95.3125) lr 0.260000 -FPS@all 820.750, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.3144 (1.3269) acc 100.0000 (97.1875) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.4988 (1.3650) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 820.740, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.5287 (1.3154) acc 93.7500 (96.8750) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.2311 (1.3451) acc 100.0000 (96.1719) lr 0.260000 -FPS@all 820.837, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:57:59 loss 1.3183 (1.2944) acc 93.7500 (97.0312) lr 0.260000 -epoch: [128/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.5662 (1.3579) acc 93.7500 (95.3906) lr 0.260000 -FPS@all 820.800, TIME@all 0.312 -epoch: [128/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:57:59 loss 1.2441 (1.2960) acc 96.8750 (97.3438) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:53 loss 1.6918 (1.3556) acc 84.3750 (95.9375) lr 0.260000 -FPS@all 820.813, TIME@all 0.312 -epoch: [128/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:57:58 loss 1.3340 (1.2959) acc 96.8750 (97.1875) lr 0.260000 -epoch: [128/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:57:53 loss 1.6742 (1.3458) acc 87.5000 (95.9375) lr 0.260000 -FPS@all 820.774, TIME@all 0.312 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:49 loss 1.2866 (1.3250) acc 100.0000 (97.5000) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4013 (1.3451) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 819.287, TIME@all 0.312 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.2991 (1.3258) acc 96.8750 (97.1875) lr 0.260000 -epoch: [129/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.3933 (1.3327) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 819.142, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.2922 (1.3251) acc 96.8750 (96.8750) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5152 (1.3377) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 819.160, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:49 loss 1.4328 (1.3236) acc 93.7500 (97.1875) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5942 (1.3470) acc 87.5000 (96.6406) lr 0.260000 -FPS@all 819.138, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:49 loss 1.2317 (1.3216) acc 96.8750 (97.1875) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5366 (1.3326) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 819.181, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:57:50 loss 1.3030 (1.3364) acc 100.0000 (96.8750) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.5083 (1.3616) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 819.172, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:49 loss 1.3842 (1.3166) acc 93.7500 (97.0312) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4849 (1.3453) acc 90.6250 (95.7812) lr 0.260000 -FPS@all 819.187, TIME@all 0.313 -epoch: [129/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:57:50 loss 1.3290 (1.3424) acc 96.8750 (96.4062) lr 0.260000 -epoch: [129/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:57:43 loss 1.4364 (1.3539) acc 93.7500 (95.7031) lr 0.260000 -FPS@all 819.185, TIME@all 0.313 -epoch: [130/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.4526 (1.3210) acc 93.7500 (96.7188) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:57:30 loss 1.3440 (1.3078) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 818.634, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:32 loss 1.2557 (1.3202) acc 96.8750 (95.7812) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:57:29 loss 1.3596 (1.3267) acc 96.8750 (96.1719) lr 0.260000 -FPS@all 818.713, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:33 loss 1.3131 (1.2905) acc 96.8750 (97.3438) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.3256 (1.2965) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 818.587, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.2130 (1.2631) acc 100.0000 (98.4375) lr 0.260000 -epoch: [130/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:57:30 loss 1.3250 (1.2945) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 818.602, TIME@all 0.313 -epoch: [130/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:57:33 loss 1.2682 (1.2927) acc 100.0000 (98.1250) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.2304 (1.3140) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 818.600, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:32 loss 1.1960 (1.2876) acc 100.0000 (97.5000) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:57:30 loss 1.3639 (1.2845) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 818.639, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:57:32 loss 1.3350 (1.2798) acc 96.8750 (97.3438) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:57:30 loss 1.2937 (1.2901) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 818.562, TIME@all 0.313 -epoch: [130/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:57:33 loss 1.3158 (1.2839) acc 93.7500 (96.8750) lr 0.260000 -epoch: [130/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:57:30 loss 1.3613 (1.2873) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 818.618, TIME@all 0.313 -epoch: [131/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:57:10 loss 1.2940 (1.3069) acc 96.8750 (96.8750) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:57:04 loss 1.2532 (1.3430) acc 96.8750 (96.2500) lr 0.260000 -FPS@all 820.678, TIME@all 0.312 -epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:12 loss 1.2310 (1.3037) acc 96.8750 (97.3438) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2454 (1.3288) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 820.401, TIME@all 0.312 -epoch: [131/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4684 (1.3167) acc 90.6250 (96.8750) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.2651 (1.3410) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 820.495, TIME@all 0.312 -epoch: [131/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.2819 (1.3243) acc 96.8750 (96.0938) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.3974 (1.3426) acc 93.7500 (96.0938) lr 0.260000 -FPS@all 820.451, TIME@all 0.312 -epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:12 loss 1.4192 (1.3218) acc 93.7500 (96.4062) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.4442 (1.3289) acc 93.7500 (96.4062) lr 0.260000 -FPS@all 820.420, TIME@all 0.312 -epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.3217 (1.3047) acc 96.8750 (97.5000) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2610 (1.3273) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.490, TIME@all 0.312 -epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4061 (1.3035) acc 93.7500 (96.4062) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:57:05 loss 1.2734 (1.3396) acc 100.0000 (96.0938) lr 0.260000 -FPS@all 820.458, TIME@all 0.312 -epoch: [131/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:57:11 loss 1.4465 (1.2992) acc 93.7500 (97.5000) lr 0.260000 -epoch: [131/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:57:05 loss 1.3393 (1.3253) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 820.509, TIME@all 0.312 -epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:57:08 loss 1.6798 (1.3224) acc 90.6250 (96.8750) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.3637 (1.3538) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 817.533, TIME@all 0.313 -epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:57:09 loss 1.5463 (1.3126) acc 87.5000 (97.5000) lr 0.260000 -epoch: [132/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:57:02 loss 1.2848 (1.3330) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 817.363, TIME@all 0.313 -epoch: [132/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.3603 (1.3091) acc 96.8750 (97.0312) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2990 (1.3238) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 817.413, TIME@all 0.313 -epoch: [132/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.4834 (1.3185) acc 90.6250 (97.0312) lr 0.260000 -epoch: [132/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:57:02 loss 1.2193 (1.3454) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 817.353, TIME@all 0.313 -epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:57:08 loss 1.4365 (1.3198) acc 90.6250 (96.8750) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:02 loss 1.4076 (1.3539) acc 90.6250 (95.8594) lr 0.260000 -FPS@all 817.372, TIME@all 0.313 -epoch: [132/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:57:08 loss 1.4646 (1.3228) acc 93.7500 (96.0938) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2791 (1.3402) acc 96.8750 (96.0938) lr 0.260000 -FPS@all 817.417, TIME@all 0.313 -epoch: [132/350][20/50] time 0.316 (0.314) data 0.001 (0.013) eta 0:57:08 loss 1.3006 (1.3230) acc 100.0000 (96.7188) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.001 (0.007) eta 0:57:01 loss 1.3082 (1.3548) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 817.387, TIME@all 0.313 -epoch: [132/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:57:08 loss 1.6315 (1.3143) acc 87.5000 (96.4062) lr 0.260000 -epoch: [132/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:57:01 loss 1.2599 (1.3417) acc 96.8750 (95.9375) lr 0.260000 -FPS@all 817.426, TIME@all 0.313 -epoch: [133/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.2503 (1.2672) acc 100.0000 (97.9688) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.4011 (1.3117) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 822.754, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.014) eta 0:56:40 loss 1.2640 (1.2802) acc 100.0000 (97.8125) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.2516 (1.2980) acc 100.0000 (97.2656) lr 0.260000 -FPS@all 822.844, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.3401 (1.3027) acc 90.6250 (97.5000) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:27 loss 1.4148 (1.3093) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 822.707, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.5672 (1.2837) acc 90.6250 (97.1875) lr 0.260000 -epoch: [133/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:56:27 loss 1.3829 (1.3078) acc 90.6250 (96.8750) lr 0.260000 -FPS@all 822.730, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:56:41 loss 1.3584 (1.2846) acc 96.8750 (98.1250) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:27 loss 1.2818 (1.3122) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 822.758, TIME@all 0.311 -epoch: [133/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:56:41 loss 1.4232 (1.2703) acc 93.7500 (98.1250) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.3546 (1.2997) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 822.731, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.000 (0.014) eta 0:56:41 loss 1.3347 (1.2721) acc 100.0000 (98.1250) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:56:26 loss 1.3018 (1.2908) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 822.768, TIME@all 0.311 -epoch: [133/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:56:41 loss 1.3286 (1.2507) acc 96.8750 (98.5938) lr 0.260000 -epoch: [133/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:56:26 loss 1.4027 (1.2871) acc 93.7500 (97.8906) lr 0.260000 -FPS@all 822.776, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4061 (1.2348) acc 93.7500 (99.0625) lr 0.260000 -epoch: [134/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2425 (1.2460) acc 100.0000 (98.9844) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.906, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:21 loss 1.1831 (1.2354) acc 100.0000 (98.1250) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1647 (1.2502) acc 100.0000 (98.0469) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.969, TIME@all 0.311 -epoch: [134/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:56:22 loss 1.3680 (1.2684) acc 96.8750 (97.5000) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2452 (1.2740) acc 90.6250 (97.1094) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.838, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.2789 (1.2306) acc 100.0000 (98.7500) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:17 loss 1.3267 (1.2643) acc 96.8750 (97.8125) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.781, TIME@all 0.312 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4153 (1.2703) acc 87.5000 (97.1875) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2740 (1.2935) acc 96.8750 (96.7969) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.849, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.3039 (1.2297) acc 93.7500 (99.0625) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.2452 (1.2470) acc 100.0000 (98.5938) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.862, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.2779 (1.2449) acc 96.8750 (98.2812) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1626 (1.2651) acc 100.0000 (97.5781) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.869, TIME@all 0.311 -epoch: [134/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:56:22 loss 1.4239 (1.2407) acc 90.6250 (98.5938) lr 0.260000 -epoch: [134/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:56:16 loss 1.1717 (1.2779) acc 100.0000 (97.1094) lr 0.260000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 821.856, TIME@all 0.311 -epoch: [135/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:56:26 loss 1.4440 (1.2633) acc 93.7500 (97.9688) lr 0.260000 -epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.2722 (1.2761) acc 96.8750 (98.1250) lr 0.260000 -FPS@all 819.169, TIME@all 0.313 -epoch: [135/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:56:26 loss 1.2236 (1.2330) acc 100.0000 (97.9688) lr 0.260000 -epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2205 (1.2618) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 819.258, TIME@all 0.312 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:56:27 loss 1.3356 (1.2560) acc 96.8750 (98.7500) lr 0.260000 -epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:11 loss 1.3273 (1.2930) acc 93.7500 (97.8125) lr 0.260000 -FPS@all 819.105, TIME@all 0.313 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:56:27 loss 1.4384 (1.2674) acc 93.7500 (97.6562) lr 0.260000 -epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.2826 (1.2810) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 819.087, TIME@all 0.313 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:56:27 loss 1.3623 (1.2796) acc 100.0000 (97.9688) lr 0.260000 -epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.5304 (1.2875) acc 90.6250 (97.8125) lr 0.260000 -FPS@all 819.110, TIME@all 0.313 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:56:27 loss 1.3413 (1.2612) acc 96.8750 (98.4375) lr 0.260000 -epoch: [135/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2375 (1.2651) acc 100.0000 (98.4375) lr 0.260000 -FPS@all 819.118, TIME@all 0.313 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:56:27 loss 1.3704 (1.2654) acc 93.7500 (97.6562) lr 0.260000 -epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:56:10 loss 1.2192 (1.2803) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 819.077, TIME@all 0.313 -epoch: [135/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:56:26 loss 1.2483 (1.2560) acc 100.0000 (97.6562) lr 0.260000 -epoch: [135/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:56:10 loss 1.1857 (1.2797) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 819.163, TIME@all 0.313 -epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.3796 (1.2795) acc 100.0000 (97.6562) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2228 (1.2892) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 820.953, TIME@all 0.312 -epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:55:47 loss 1.3847 (1.2934) acc 93.7500 (97.0312) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:55:40 loss 1.3094 (1.3011) acc 93.7500 (96.5625) lr 0.260000 -FPS@all 820.838, TIME@all 0.312 -epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:55:47 loss 1.4261 (1.2493) acc 90.6250 (97.8125) lr 0.260000 -epoch: [136/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:55:41 loss 1.2702 (1.2900) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 820.821, TIME@all 0.312 -epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:48 loss 1.2572 (1.2831) acc 93.7500 (97.3438) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2319 (1.3100) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 820.854, TIME@all 0.312 -epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.3041 (1.2769) acc 96.8750 (97.9688) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2741 (1.2914) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 820.855, TIME@all 0.312 -epoch: [136/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2717 (1.2795) acc 100.0000 (97.5000) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2664 (1.3130) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 820.890, TIME@all 0.312 -epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2139 (1.2762) acc 100.0000 (97.3438) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.2881 (1.3053) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 820.909, TIME@all 0.312 -epoch: [136/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:47 loss 1.2553 (1.2722) acc 96.8750 (98.4375) lr 0.260000 -epoch: [136/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:55:40 loss 1.1817 (1.2853) acc 100.0000 (97.7344) lr 0.260000 -FPS@all 820.862, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.2248 (1.2564) acc 100.0000 (99.0625) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.1623 (1.2853) acc 100.0000 (98.1250) lr 0.260000 -FPS@all 821.770, TIME@all 0.312 -epoch: [137/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:55:32 loss 1.2972 (1.2748) acc 96.8750 (98.1250) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.007) eta 0:55:25 loss 1.3615 (1.3033) acc 90.6250 (97.6562) lr 0.260000 -FPS@all 821.852, TIME@all 0.311 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3347 (1.2436) acc 96.8750 (99.0625) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.4410 (1.2931) acc 87.5000 (97.3438) lr 0.260000 -FPS@all 821.678, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3103 (1.2697) acc 96.8750 (97.5000) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2866 (1.3036) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 821.670, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.3784 (1.2731) acc 93.7500 (98.2812) lr 0.260000 -epoch: [137/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.3210 (1.3030) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 821.678, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:32 loss 1.4110 (1.2848) acc 93.7500 (97.3438) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2853 (1.3098) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.741, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.1759 (1.2557) acc 100.0000 (98.7500) lr 0.260000 -epoch: [137/350][40/50] time 0.308 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2404 (1.2819) acc 100.0000 (98.0469) lr 0.260000 -FPS@all 821.711, TIME@all 0.312 -epoch: [137/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:55:33 loss 1.4330 (1.2849) acc 93.7500 (97.8125) lr 0.260000 -epoch: [137/350][40/50] time 0.307 (0.312) data 0.000 (0.006) eta 0:55:26 loss 1.2648 (1.3102) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 821.724, TIME@all 0.312 -epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:55:23 loss 1.2661 (1.2619) acc 96.8750 (98.2812) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3513 (1.3061) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 821.387, TIME@all 0.312 -epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2147 (1.2722) acc 100.0000 (97.5000) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3693 (1.3188) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 821.433, TIME@all 0.312 -epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2965 (1.2864) acc 96.8750 (97.3438) lr 0.260000 -epoch: [138/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.4011 (1.3184) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 821.272, TIME@all 0.312 -epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:55:23 loss 1.1720 (1.2680) acc 100.0000 (98.2812) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:55:14 loss 1.2676 (1.2892) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 821.277, TIME@all 0.312 -epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.4243 (1.2878) acc 96.8750 (97.8125) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3302 (1.3097) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 821.329, TIME@all 0.312 -epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.4315 (1.2970) acc 96.8750 (96.8750) lr 0.260000 -epoch: [138/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.3129 (1.3071) acc 96.8750 (96.3281) lr 0.260000 -FPS@all 821.367, TIME@all 0.312 -epoch: [138/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2118 (1.2753) acc 100.0000 (97.6562) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.2226 (1.3002) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 821.290, TIME@all 0.312 -epoch: [138/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:23 loss 1.2020 (1.2711) acc 100.0000 (98.5938) lr 0.260000 -epoch: [138/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:55:14 loss 1.5243 (1.3016) acc 87.5000 (97.3438) lr 0.260000 -FPS@all 821.326, TIME@all 0.312 -epoch: [139/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:55:12 loss 1.5441 (1.3354) acc 93.7500 (96.0938) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5793 (1.3677) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 819.198, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:10 loss 1.4881 (1.3068) acc 96.8750 (97.5000) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5108 (1.3383) acc 90.6250 (96.6406) lr 0.260000 -FPS@all 819.298, TIME@all 0.312 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.3762 (1.3419) acc 96.8750 (96.5625) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.3505 (1.3769) acc 90.6250 (95.8594) lr 0.260000 -FPS@all 819.133, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:55:10 loss 1.4094 (1.3288) acc 96.8750 (96.2500) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:55:06 loss 1.4796 (1.3511) acc 93.7500 (96.0156) lr 0.260000 -FPS@all 819.148, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:55:11 loss 1.4845 (1.3660) acc 93.7500 (96.2500) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.4204 (1.3805) acc 93.7500 (95.6250) lr 0.260000 -FPS@all 819.154, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.5491 (1.3669) acc 93.7500 (95.7812) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.3355 (1.3779) acc 93.7500 (95.9375) lr 0.260000 -FPS@all 819.166, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:55:11 loss 1.4810 (1.3419) acc 96.8750 (96.4062) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:55:06 loss 1.3898 (1.3614) acc 100.0000 (95.9375) lr 0.260000 -FPS@all 819.183, TIME@all 0.313 -epoch: [139/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:55:11 loss 1.4743 (1.3438) acc 96.8750 (95.9375) lr 0.260000 -epoch: [139/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:55:06 loss 1.5403 (1.3717) acc 90.6250 (95.6250) lr 0.260000 -FPS@all 819.174, TIME@all 0.313 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.4470 (1.3174) acc 96.8750 (96.7188) lr 0.260000 -epoch: [140/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:54:58 loss 1.2609 (1.3133) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 816.357, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3200 (1.2924) acc 96.8750 (97.6562) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:58 loss 1.2149 (1.2937) acc 100.0000 (97.4219) lr 0.260000 -FPS@all 816.422, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.3681 (1.3100) acc 93.7500 (97.0312) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.2295 (1.3138) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 816.266, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.3224 (1.2879) acc 96.8750 (97.8125) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3063 (1.3067) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 816.239, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3472 (1.2975) acc 96.8750 (97.3438) lr 0.260000 -epoch: [140/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3434 (1.3054) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 816.338, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:02 loss 1.5682 (1.3206) acc 96.8750 (97.3438) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.2570 (1.3170) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 816.320, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3718 (1.2795) acc 96.8750 (98.1250) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3270 (1.2799) acc 93.7500 (97.9688) lr 0.260000 -FPS@all 816.329, TIME@all 0.314 -epoch: [140/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:55:01 loss 1.3783 (1.2846) acc 96.8750 (98.4375) lr 0.260000 -epoch: [140/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:54:59 loss 1.3377 (1.2966) acc 93.7500 (97.8125) lr 0.260000 -FPS@all 816.308, TIME@all 0.314 -epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.2899 (1.2886) acc 96.8750 (97.9688) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.3957 (1.3138) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 819.780, TIME@all 0.312 -epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3784 (1.2635) acc 96.8750 (98.2812) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4354 (1.2939) acc 93.7500 (98.0469) lr 0.260000 -FPS@all 819.631, TIME@all 0.312 -epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3668 (1.2953) acc 96.8750 (97.1875) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4462 (1.3207) acc 93.7500 (96.2500) lr 0.260000 -FPS@all 819.680, TIME@all 0.312 -epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3688 (1.2935) acc 96.8750 (97.6562) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:54:33 loss 1.3553 (1.3390) acc 96.8750 (96.0156) lr 0.260000 -FPS@all 819.643, TIME@all 0.312 -epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:54:50 loss 1.3202 (1.2968) acc 96.8750 (97.8125) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:54:33 loss 1.8302 (1.3142) acc 84.3750 (97.1094) lr 0.260000 -FPS@all 819.626, TIME@all 0.312 -epoch: [141/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.4627 (1.2750) acc 93.7500 (98.4375) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.4423 (1.3153) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 819.654, TIME@all 0.312 -epoch: [141/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.3009 (1.2794) acc 96.8750 (96.7188) lr 0.260000 -epoch: [141/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.5243 (1.3163) acc 90.6250 (96.3281) lr 0.260000 -FPS@all 819.638, TIME@all 0.312 -epoch: [141/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:54:50 loss 1.4179 (1.2853) acc 96.8750 (97.0312) lr 0.260000 -epoch: [141/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:54:33 loss 1.3025 (1.3102) acc 96.8750 (96.5625) lr 0.260000 -FPS@all 819.645, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.3344 (1.3175) acc 96.8750 (96.5625) lr 0.260000 -epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2469 (1.3253) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 821.526, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.3265 (1.3308) acc 93.7500 (96.5625) lr 0.260000 -epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2549 (1.3336) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 821.552, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.2971 (1.3132) acc 96.8750 (96.4062) lr 0.260000 -epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2757 (1.3272) acc 93.7500 (95.8594) lr 0.260000 -FPS@all 821.492, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:54:24 loss 1.4382 (1.3251) acc 93.7500 (97.1875) lr 0.260000 -epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:54:10 loss 1.2840 (1.3243) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 821.433, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:54:24 loss 1.4371 (1.3341) acc 96.8750 (97.0312) lr 0.260000 -epoch: [142/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:54:10 loss 1.3809 (1.3330) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 821.438, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.3842 (1.3421) acc 90.6250 (95.7812) lr 0.260000 -epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.3971 (1.3238) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 821.462, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:23 loss 1.2558 (1.3138) acc 100.0000 (97.9688) lr 0.260000 -epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2348 (1.3293) acc 100.0000 (96.7969) lr 0.260000 -FPS@all 821.457, TIME@all 0.312 -epoch: [142/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:54:24 loss 1.3104 (1.3142) acc 96.8750 (97.5000) lr 0.260000 -epoch: [142/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:54:10 loss 1.2873 (1.3235) acc 96.8750 (97.1875) lr 0.260000 -FPS@all 821.474, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3324 (1.2590) acc 96.8750 (98.1250) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2632 (1.2771) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 819.891, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:10 loss 1.3047 (1.2866) acc 93.7500 (97.5000) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2220 (1.3007) acc 100.0000 (97.6562) lr 0.260000 -FPS@all 819.970, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.2664 (1.2763) acc 96.8750 (97.8125) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:01 loss 1.3104 (1.3095) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 819.785, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:54:11 loss 1.2360 (1.2685) acc 100.0000 (97.9688) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:54:01 loss 1.3476 (1.3002) acc 93.7500 (97.2656) lr 0.260000 -FPS@all 819.804, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3154 (1.2767) acc 93.7500 (97.9688) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2119 (1.2934) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 819.829, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:54:11 loss 1.3129 (1.2645) acc 100.0000 (98.2812) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2200 (1.2885) acc 100.0000 (96.9531) lr 0.260000 -FPS@all 819.837, TIME@all 0.312 -epoch: [143/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:54:11 loss 1.3177 (1.2878) acc 100.0000 (97.6562) lr 0.260000 -epoch: [143/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:54:00 loss 1.2783 (1.2949) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 819.851, TIME@all 0.312 -epoch: [143/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:54:11 loss 1.2512 (1.2580) acc 100.0000 (97.8125) lr 0.260000 -epoch: [143/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:54:01 loss 1.2167 (1.3013) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 819.846, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:53:51 loss 1.2800 (1.2813) acc 96.8750 (97.0312) lr 0.260000 -epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3164 (1.3073) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 819.560, TIME@all 0.312 -epoch: [144/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.1827 (1.2746) acc 100.0000 (97.6562) lr 0.260000 -epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3127 (1.3066) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 819.515, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:52 loss 1.3411 (1.2912) acc 96.8750 (97.0312) lr 0.260000 -epoch: [144/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:53:41 loss 1.3074 (1.3049) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 819.395, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:53:52 loss 1.2172 (1.2805) acc 96.8750 (96.4062) lr 0.260000 -epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:53:41 loss 1.3115 (1.3032) acc 96.8750 (96.8750) lr 0.260000 -FPS@all 819.413, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:53:52 loss 1.2686 (1.2754) acc 93.7500 (96.4062) lr 0.260000 -epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2814 (1.2904) acc 100.0000 (96.6406) lr 0.260000 -FPS@all 819.456, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.1580 (1.2807) acc 100.0000 (97.8125) lr 0.260000 -epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.3171 (1.3010) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 819.444, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:51 loss 1.3047 (1.2778) acc 100.0000 (97.8125) lr 0.260000 -epoch: [144/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2144 (1.2989) acc 100.0000 (97.1094) lr 0.260000 -FPS@all 819.506, TIME@all 0.312 -epoch: [144/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:52 loss 1.2027 (1.2650) acc 100.0000 (97.5000) lr 0.260000 -epoch: [144/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:53:41 loss 1.2080 (1.2904) acc 96.8750 (96.7969) lr 0.260000 -FPS@all 819.464, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3357 (1.2578) acc 96.8750 (97.9688) lr 0.260000 -epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.4115 (1.2876) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 821.037, TIME@all 0.312 -epoch: [145/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:53:31 loss 1.1971 (1.2327) acc 100.0000 (98.4375) lr 0.260000 -epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:53:24 loss 1.3621 (1.2683) acc 96.8750 (97.4219) lr 0.260000 -FPS@all 821.054, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:32 loss 1.2022 (1.2720) acc 100.0000 (98.1250) lr 0.260000 -epoch: [145/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:53:25 loss 1.4300 (1.2932) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 820.903, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.2889 (1.2500) acc 96.8750 (97.8125) lr 0.260000 -epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.6729 (1.2756) acc 87.5000 (97.4219) lr 0.260000 -FPS@all 820.983, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.001 (0.013) eta 0:53:32 loss 1.2795 (1.2651) acc 93.7500 (97.3438) lr 0.260000 -epoch: [145/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:53:24 loss 1.6089 (1.2895) acc 84.3750 (97.3438) lr 0.260000 -FPS@all 820.934, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3493 (1.2587) acc 90.6250 (97.1875) lr 0.260000 -epoch: [145/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:53:24 loss 1.4169 (1.2770) acc 96.8750 (97.5781) lr 0.260000 -FPS@all 820.918, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.4527 (1.2534) acc 93.7500 (98.1250) lr 0.260000 -epoch: [145/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:53:24 loss 1.3442 (1.2835) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 821.003, TIME@all 0.312 -epoch: [145/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:53:31 loss 1.3707 (1.2722) acc 96.8750 (98.4375) lr 0.260000 -epoch: [145/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:53:24 loss 1.3266 (1.2810) acc 96.8750 (97.9688) lr 0.260000 -FPS@all 820.981, TIME@all 0.312 -epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:19 loss 1.2036 (1.2608) acc 100.0000 (97.5000) lr 0.260000 -epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.1901 (1.3021) acc 100.0000 (96.4062) lr 0.260000 -FPS@all 819.697, TIME@all 0.312 -epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:19 loss 1.4337 (1.2988) acc 93.7500 (97.5000) lr 0.260000 -epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2630 (1.3346) acc 96.8750 (96.4844) lr 0.260000 -FPS@all 819.687, TIME@all 0.312 -epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.3097 (1.2826) acc 100.0000 (97.8125) lr 0.260000 -epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2493 (1.3010) acc 93.7500 (97.4219) lr 0.260000 -FPS@all 819.552, TIME@all 0.312 -epoch: [146/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:53:20 loss 1.4727 (1.2797) acc 93.7500 (98.1250) lr 0.260000 -epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:53:15 loss 1.3330 (1.2966) acc 96.8750 (97.8125) lr 0.260000 -FPS@all 819.588, TIME@all 0.312 -epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.4380 (1.2984) acc 96.8750 (97.3438) lr 0.260000 -epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2718 (1.3137) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 819.648, TIME@all 0.312 -epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.5463 (1.2822) acc 87.5000 (98.1250) lr 0.260000 -epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.2692 (1.3055) acc 96.8750 (97.6562) lr 0.260000 -FPS@all 819.623, TIME@all 0.312 -epoch: [146/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.2797 (1.2993) acc 93.7500 (96.8750) lr 0.260000 -epoch: [146/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.3690 (1.2995) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 819.601, TIME@all 0.312 -epoch: [146/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:53:20 loss 1.3978 (1.3016) acc 93.7500 (96.8750) lr 0.260000 -epoch: [146/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:53:15 loss 1.3959 (1.3105) acc 93.7500 (96.7188) lr 0.260000 -FPS@all 819.429, TIME@all 0.312 -epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.3507 (1.2709) acc 100.0000 (97.5000) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.2979 (1.3085) acc 96.8750 (97.1094) lr 0.260000 -FPS@all 819.343, TIME@all 0.312 -epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.3864 (1.2765) acc 96.8750 (98.2812) lr 0.260000 -epoch: [147/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.2503 (1.2850) acc 93.7500 (97.7344) lr 0.260000 -FPS@all 819.404, TIME@all 0.312 -epoch: [147/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:53:01 loss 1.3498 (1.2768) acc 93.7500 (97.9688) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:52:55 loss 1.2707 (1.2927) acc 96.8750 (97.2656) lr 0.260000 -FPS@all 819.234, TIME@all 0.312 -epoch: [147/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:53:01 loss 1.2052 (1.2894) acc 100.0000 (96.8750) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:52:55 loss 1.3789 (1.3051) acc 96.8750 (97.0312) lr 0.260000 -FPS@all 819.270, TIME@all 0.312 -epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.3381 (1.2547) acc 96.8750 (98.4375) lr 0.260000 -epoch: [147/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.2865 (1.2942) acc 93.7500 (97.0312) lr 0.260000 -FPS@all 819.342, TIME@all 0.312 -epoch: [147/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:53:01 loss 1.2945 (1.2495) acc 96.8750 (98.2812) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:52:54 loss 1.1901 (1.2903) acc 100.0000 (97.1875) lr 0.260000 -FPS@all 819.339, TIME@all 0.312 -epoch: [147/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.3316 (1.2722) acc 96.8750 (97.1875) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.3496 (1.2924) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 819.344, TIME@all 0.312 -epoch: [147/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:53:00 loss 1.4370 (1.2872) acc 93.7500 (97.0312) lr 0.260000 -epoch: [147/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:52:55 loss 1.2425 (1.3080) acc 100.0000 (96.7188) lr 0.260000 -FPS@all 819.294, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:52:40 loss 1.2006 (1.2685) acc 100.0000 (97.5000) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:35 loss 1.4368 (1.3026) acc 96.8750 (96.6406) lr 0.260000 -FPS@all 821.766, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.3127 (1.2767) acc 96.8750 (98.1250) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3369 (1.3154) acc 93.7500 (96.7969) lr 0.260000 -FPS@all 821.650, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.3071 (1.3023) acc 100.0000 (97.6562) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.4755 (1.3248) acc 90.6250 (96.9531) lr 0.260000 -FPS@all 821.582, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:42 loss 1.4044 (1.2865) acc 96.8750 (97.8125) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.5110 (1.3181) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 821.562, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4168 (1.2882) acc 90.6250 (97.6562) lr 0.260000 -epoch: [148/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3506 (1.3442) acc 93.7500 (96.6406) lr 0.260000 -FPS@all 821.661, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4677 (1.2808) acc 93.7500 (98.1250) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3924 (1.3149) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 821.581, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:52:41 loss 1.4685 (1.2995) acc 93.7500 (97.3438) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.5286 (1.3141) acc 90.6250 (97.0312) lr 0.260000 -FPS@all 821.606, TIME@all 0.312 -epoch: [148/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:52:41 loss 1.4934 (1.3033) acc 90.6250 (96.5625) lr 0.260000 -epoch: [148/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:52:36 loss 1.3794 (1.3334) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 821.605, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.3518 (1.3037) acc 96.8750 (97.0312) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.4412 (1.3119) acc 90.6250 (96.5625) lr 0.260000 -FPS@all 820.226, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:52:31 loss 1.2645 (1.2729) acc 100.0000 (97.5000) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:52:23 loss 1.4622 (1.2994) acc 93.7500 (97.1094) lr 0.260000 -FPS@all 820.133, TIME@all 0.312 -epoch: [149/350][20/50] time 0.309 (0.313) data 0.000 (0.014) eta 0:52:30 loss 1.2608 (1.2959) acc 96.8750 (97.5000) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3316 (1.2930) acc 93.7500 (97.3438) lr 0.260000 -FPS@all 820.234, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.2324 (1.3195) acc 96.8750 (96.8750) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3053 (1.3279) acc 100.0000 (96.4844) lr 0.260000 -FPS@all 820.151, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:31 loss 1.2632 (1.2636) acc 100.0000 (99.2188) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.2418 (1.2883) acc 100.0000 (98.2812) lr 0.260000 -FPS@all 820.176, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:31 loss 1.2376 (1.2727) acc 96.8750 (97.6562) lr 0.260000 -epoch: [149/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:52:23 loss 1.4074 (1.2908) acc 96.8750 (96.9531) lr 0.260000 -FPS@all 820.150, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.3735 (1.2807) acc 93.7500 (96.5625) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.3156 (1.3001) acc 96.8750 (96.4062) lr 0.260000 -FPS@all 820.138, TIME@all 0.312 -epoch: [149/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:52:30 loss 1.1852 (1.2862) acc 100.0000 (97.3438) lr 0.260000 -epoch: [149/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:52:23 loss 1.2744 (1.2981) acc 100.0000 (97.5000) lr 0.260000 -FPS@all 820.182, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:13 loss 1.2652 (1.2369) acc 100.0000 (98.5938) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:10 loss 1.2839 (1.2688) acc 100.0000 (97.8906) lr 0.260000 -FPS@all 819.938, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3262 (1.2410) acc 96.8750 (98.5938) lr 0.260000 -epoch: [150/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3729 (1.3000) acc 93.7500 (96.9531) lr 0.260000 -FPS@all 819.821, TIME@all 0.312 -epoch: [150/350][20/50] time 0.306 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.5075 (1.2949) acc 90.6250 (97.1875) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3337 (1.3147) acc 93.7500 (96.3281) lr 0.260000 -FPS@all 819.727, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3045 (1.2773) acc 96.8750 (98.1250) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.4422 (1.3003) acc 93.7500 (97.1875) lr 0.260000 -FPS@all 819.800, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3580 (1.2617) acc 90.6250 (98.2812) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3714 (1.3012) acc 96.8750 (97.5000) lr 0.260000 -FPS@all 819.716, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.3207 (1.2793) acc 100.0000 (97.5000) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.5099 (1.3022) acc 90.6250 (97.2656) lr 0.260000 -FPS@all 819.752, TIME@all 0.312 -epoch: [150/350][20/50] time 0.307 (0.312) data 0.000 (0.012) eta 0:52:14 loss 1.2934 (1.2575) acc 100.0000 (97.9688) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:52:11 loss 1.3508 (1.2848) acc 96.8750 (97.3438) lr 0.260000 -FPS@all 819.566, TIME@all 0.312 -epoch: [150/350][20/50] time 0.306 (0.313) data 0.000 (0.013) eta 0:52:14 loss 1.2708 (1.2654) acc 100.0000 (98.2812) lr 0.260000 -epoch: [150/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:52:11 loss 1.3161 (1.3001) acc 93.7500 (97.5781) lr 0.260000 -FPS@all 819.762, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1757 (1.2032) acc 100.0000 (98.7500) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1901 (1.2070) acc 100.0000 (98.6719) lr 0.026000 -FPS@all 819.772, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:51:55 loss 1.1401 (1.2007) acc 100.0000 (99.3750) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:50 loss 1.2448 (1.2158) acc 96.8750 (98.6719) lr 0.026000 -FPS@all 819.672, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:51:54 loss 1.2299 (1.2158) acc 96.8750 (98.5938) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.2319 (1.2227) acc 100.0000 (98.4375) lr 0.026000 -FPS@all 819.602, TIME@all 0.312 -epoch: [151/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:51:54 loss 1.1744 (1.1974) acc 100.0000 (99.6875) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.1756 (1.2157) acc 96.8750 (98.5938) lr 0.026000 -FPS@all 819.583, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:51:55 loss 1.1654 (1.2024) acc 96.8750 (98.9062) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:51:51 loss 1.1466 (1.2036) acc 100.0000 (98.5938) lr 0.026000 -FPS@all 819.635, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:55 loss 1.1773 (1.2070) acc 100.0000 (99.3750) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:51 loss 1.1540 (1.2164) acc 100.0000 (98.7500) lr 0.026000 -FPS@all 819.616, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1788 (1.2123) acc 100.0000 (99.0625) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1278 (1.2023) acc 100.0000 (99.0625) lr 0.026000 -FPS@all 819.683, TIME@all 0.312 -epoch: [151/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:51:54 loss 1.1732 (1.2140) acc 100.0000 (98.7500) lr 0.026000 -epoch: [151/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:50 loss 1.1512 (1.2099) acc 100.0000 (98.5938) lr 0.026000 -FPS@all 819.672, TIME@all 0.312 -epoch: [152/350][20/50] time 0.319 (0.315) data 0.001 (0.014) eta 0:52:06 loss 1.1614 (1.1623) acc 100.0000 (99.5312) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1513 (1.1616) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 815.327, TIME@all 0.314 -epoch: [152/350][20/50] time 0.319 (0.315) data 0.000 (0.012) eta 0:52:06 loss 1.1728 (1.1479) acc 100.0000 (99.8438) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:52:04 loss 1.1456 (1.1587) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 815.184, TIME@all 0.314 -epoch: [152/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:52:06 loss 1.2145 (1.1579) acc 96.8750 (99.3750) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1283 (1.1551) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 815.185, TIME@all 0.314 -epoch: [152/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:52:06 loss 1.2327 (1.1468) acc 96.8750 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1239 (1.1442) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 815.249, TIME@all 0.314 -epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1400 (1.1514) acc 100.0000 (99.5312) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1615 (1.1568) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 815.239, TIME@all 0.314 -epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1561 (1.1541) acc 100.0000 (99.6875) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1175 (1.1533) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 815.190, TIME@all 0.314 -epoch: [152/350][20/50] time 0.319 (0.315) data 0.001 (0.013) eta 0:52:06 loss 1.1795 (1.1524) acc 100.0000 (99.3750) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:04 loss 1.1390 (1.1513) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 815.208, TIME@all 0.314 -epoch: [152/350][20/50] time 0.318 (0.315) data 0.001 (0.013) eta 0:52:05 loss 1.4165 (1.1698) acc 93.7500 (99.5312) lr 0.026000 -epoch: [152/350][40/50] time 0.313 (0.315) data 0.000 (0.007) eta 0:52:03 loss 1.1470 (1.1663) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 815.362, TIME@all 0.314 -epoch: [153/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 0:51:25 loss 1.1228 (1.1245) acc 100.0000 (100.0000) lr 0.026000 -epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:19 loss 1.1777 (1.1345) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 821.441, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1448 (1.1319) acc 96.8750 (99.3750) lr 0.026000 -epoch: [153/350][40/50] time 0.313 (0.312) data 0.001 (0.006) eta 0:51:20 loss 1.1731 (1.1454) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 821.278, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1854 (1.1363) acc 96.8750 (99.6875) lr 0.026000 -epoch: [153/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1384 (1.1390) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 821.305, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1107 (1.1264) acc 100.0000 (100.0000) lr 0.026000 -epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1159 (1.1394) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 821.315, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1369 (1.1244) acc 100.0000 (100.0000) lr 0.026000 -epoch: [153/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.1371 (1.1439) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 821.319, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1170 (1.1328) acc 100.0000 (99.6875) lr 0.026000 -epoch: [153/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:51:20 loss 1.1460 (1.1408) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 821.327, TIME@all 0.312 -epoch: [153/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:51:26 loss 1.2284 (1.1283) acc 96.8750 (99.8438) lr 0.026000 -epoch: [153/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:51:20 loss 1.2121 (1.1514) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 821.320, TIME@all 0.312 -epoch: [153/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:51:26 loss 1.1057 (1.1293) acc 100.0000 (99.5312) lr 0.026000 -epoch: [153/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 0:51:20 loss 1.1188 (1.1376) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 821.304, TIME@all 0.312 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:13 loss 1.1452 (1.1204) acc 100.0000 (100.0000) lr 0.026000 -epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:22 loss 1.2816 (1.1319) acc 93.7500 (99.5312) lr 0.026000 -FPS@all 816.662, TIME@all 0.313 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1021 (1.1252) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1223 (1.1285) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.452, TIME@all 0.314 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:13 loss 1.1103 (1.1200) acc 100.0000 (100.0000) lr 0.026000 -epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1593 (1.1250) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 816.529, TIME@all 0.314 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1408 (1.1135) acc 100.0000 (100.0000) lr 0.026000 -epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1953 (1.1315) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.432, TIME@all 0.314 -epoch: [154/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1309 (1.1179) acc 100.0000 (100.0000) lr 0.026000 -epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1856 (1.1357) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 816.517, TIME@all 0.314 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1020 (1.1141) acc 100.0000 (99.8438) lr 0.026000 -epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1251 (1.1313) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 816.512, TIME@all 0.314 -epoch: [154/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:51:14 loss 1.1552 (1.1240) acc 100.0000 (99.6875) lr 0.026000 -epoch: [154/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:51:23 loss 1.1250 (1.1293) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 816.470, TIME@all 0.314 -epoch: [154/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:51:14 loss 1.1164 (1.1211) acc 100.0000 (99.8438) lr 0.026000 -epoch: [154/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:51:23 loss 1.1266 (1.1269) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 816.553, TIME@all 0.314 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.2142 (1.1308) acc 96.8750 (99.3750) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1575 (1.1425) acc 100.0000 (99.1406) lr 0.026000 -FPS@all 819.568, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:50:55 loss 1.2218 (1.1237) acc 96.8750 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1317 (1.1385) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 819.645, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:56 loss 1.2237 (1.1162) acc 96.8750 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1232 (1.1280) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 819.478, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.1579 (1.1180) acc 100.0000 (100.0000) lr 0.026000 -epoch: [155/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.0979 (1.1407) acc 100.0000 (99.2969) lr 0.026000 -FPS@all 819.535, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:55 loss 1.2122 (1.1182) acc 96.8750 (99.8438) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:50:54 loss 1.1190 (1.1343) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.554, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:50:56 loss 1.1750 (1.1279) acc 100.0000 (99.6875) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1074 (1.1410) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 819.525, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:50:56 loss 1.1491 (1.1253) acc 100.0000 (99.8438) lr 0.026000 -epoch: [155/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:50:54 loss 1.1308 (1.1407) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.471, TIME@all 0.312 -epoch: [155/350][20/50] time 0.314 (0.312) data 0.001 (0.013) eta 0:50:56 loss 1.1679 (1.1121) acc 100.0000 (99.8438) lr 0.026000 -epoch: [155/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:50:54 loss 1.1229 (1.1315) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 819.494, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:37 loss 1.1063 (1.1156) acc 100.0000 (99.3750) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:50:35 loss 1.2050 (1.1293) acc 96.8750 (99.2188) lr 0.026000 -FPS@all 820.623, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.1373 (1.1091) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1184 (1.1207) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.564, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0857 (1.1049) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:36 loss 1.1245 (1.1233) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.429, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:37 loss 1.1230 (1.1117) acc 100.0000 (99.6875) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1255 (1.1317) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 820.547, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:50:38 loss 1.1414 (1.1042) acc 100.0000 (100.0000) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:50:36 loss 1.1018 (1.1167) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.476, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0908 (1.1091) acc 100.0000 (99.6875) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:36 loss 1.1560 (1.1243) acc 96.8750 (99.2969) lr 0.026000 -FPS@all 820.534, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0933 (1.1107) acc 100.0000 (99.6875) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:50:35 loss 1.1167 (1.1229) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.559, TIME@all 0.312 -epoch: [156/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:50:38 loss 1.0908 (1.1108) acc 100.0000 (99.8438) lr 0.026000 -epoch: [156/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:50:36 loss 1.1442 (1.1207) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.513, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1583 (1.1052) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.0972 (1.1151) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.911, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1054 (1.1075) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1198 (1.1135) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.802, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1080 (1.1073) acc 100.0000 (99.8438) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1094 (1.1169) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 820.813, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.0880 (1.0970) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1190 (1.1119) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.809, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:50:21 loss 1.1146 (1.1059) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:50:18 loss 1.1003 (1.1115) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.739, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1199 (1.1049) acc 100.0000 (99.6875) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.0996 (1.1072) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.808, TIME@all 0.312 -epoch: [157/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1047 (1.1038) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1022 (1.1164) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.775, TIME@all 0.312 -epoch: [157/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:50:21 loss 1.1010 (1.0985) acc 100.0000 (100.0000) lr 0.026000 -epoch: [157/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:50:18 loss 1.1010 (1.1164) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.814, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1387 (1.1080) acc 100.0000 (99.8438) lr 0.026000 -epoch: [158/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1457 (1.1161) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 820.871, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2187 (1.1095) acc 96.8750 (99.6875) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.1735 (1.1186) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.772, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1512 (1.1103) acc 100.0000 (99.6875) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1036 (1.1234) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 820.846, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2896 (1.1123) acc 93.7500 (99.6875) lr 0.026000 -epoch: [158/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.0897 (1.1172) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.715, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.1998 (1.1053) acc 100.0000 (100.0000) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:02 loss 1.1093 (1.1105) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.765, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1775 (1.1082) acc 100.0000 (100.0000) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:01 loss 1.1041 (1.1160) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.801, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:50:12 loss 1.2224 (1.1104) acc 96.8750 (99.5312) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:50:01 loss 1.1017 (1.1183) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 820.817, TIME@all 0.312 -epoch: [158/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:50:12 loss 1.1557 (1.1040) acc 100.0000 (99.8438) lr 0.026000 -epoch: [158/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:50:01 loss 1.1157 (1.1190) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.814, TIME@all 0.312 -epoch: [159/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1042 (1.0983) acc 100.0000 (99.8438) lr 0.026000 -epoch: [159/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:49:47 loss 1.1180 (1.1091) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 820.505, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.1434 (1.1019) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:49:47 loss 1.1191 (1.1073) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.506, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1227 (1.0963) acc 100.0000 (99.8438) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1407 (1.1095) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.396, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1277 (1.1006) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1316 (1.1082) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 820.405, TIME@all 0.312 -epoch: [159/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:53 loss 1.1236 (1.1023) acc 100.0000 (99.8438) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1739 (1.1087) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.439, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.0924 (1.1039) acc 100.0000 (99.8438) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:47 loss 1.1499 (1.1067) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.445, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:54 loss 1.1167 (1.1010) acc 100.0000 (99.6875) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:49:47 loss 1.1396 (1.1055) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.424, TIME@all 0.312 -epoch: [159/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:53 loss 1.0859 (1.1042) acc 100.0000 (100.0000) lr 0.026000 -epoch: [159/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:47 loss 1.1506 (1.1101) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.407, TIME@all 0.312 -epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0852 (1.0966) acc 100.0000 (99.6875) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0841 (1.1123) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 813.442, TIME@all 0.315 -epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0847 (1.1012) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0880 (1.1167) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 813.475, TIME@all 0.315 -epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.011) eta 0:50:29 loss 1.1003 (1.0911) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.311 (0.316) data 0.000 (0.006) eta 0:50:03 loss 1.0948 (1.1092) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 813.532, TIME@all 0.315 -epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1033 (1.0975) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.0855 (1.1089) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 813.325, TIME@all 0.315 -epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0974 (1.0908) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.1063 (1.1045) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 813.392, TIME@all 0.315 -epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1040 (1.0936) acc 100.0000 (99.8438) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.001 (0.006) eta 0:50:04 loss 1.0997 (1.1131) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 813.405, TIME@all 0.315 -epoch: [160/350][20/50] time 0.313 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.1075 (1.0975) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.000 (0.006) eta 0:50:04 loss 1.0893 (1.1065) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 813.366, TIME@all 0.315 -epoch: [160/350][20/50] time 0.314 (0.318) data 0.000 (0.012) eta 0:50:30 loss 1.0956 (1.0897) acc 100.0000 (100.0000) lr 0.026000 -epoch: [160/350][40/50] time 0.312 (0.316) data 0.001 (0.006) eta 0:50:04 loss 1.0797 (1.1041) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 813.373, TIME@all 0.315 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.013) eta 0:49:57 loss 1.0921 (1.1046) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.007) eta 0:50:22 loss 1.1008 (1.1196) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 805.963, TIME@all 0.318 -epoch: [161/350][20/50] time 0.334 (0.316) data 0.000 (0.013) eta 0:49:57 loss 1.0952 (1.0993) acc 100.0000 (99.5312) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.007) eta 0:50:21 loss 1.0895 (1.1082) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 806.035, TIME@all 0.318 -epoch: [161/350][20/50] time 0.334 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.1328 (1.0978) acc 100.0000 (100.0000) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.320) data 0.000 (0.006) eta 0:50:22 loss 1.0729 (1.1080) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 805.886, TIME@all 0.318 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:56 loss 1.0945 (1.0965) acc 100.0000 (100.0000) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.0996 (1.1074) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 806.148, TIME@all 0.318 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.001 (0.012) eta 0:49:57 loss 1.0965 (1.1041) acc 100.0000 (99.2188) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.0897 (1.1108) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 806.199, TIME@all 0.318 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.001 (0.012) eta 0:49:57 loss 1.1205 (1.1057) acc 100.0000 (99.3750) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.001 (0.006) eta 0:50:21 loss 1.1138 (1.1127) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 806.020, TIME@all 0.318 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.0876 (1.0977) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:22 loss 1.0879 (1.1084) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 805.946, TIME@all 0.318 -epoch: [161/350][20/50] time 0.333 (0.316) data 0.000 (0.012) eta 0:49:57 loss 1.0914 (1.0956) acc 100.0000 (99.8438) lr 0.026000 -epoch: [161/350][40/50] time 0.311 (0.319) data 0.000 (0.006) eta 0:50:21 loss 1.1010 (1.1103) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 806.027, TIME@all 0.318 -epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.0997 (1.0897) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1382 (1.1014) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.297, TIME@all 0.312 -epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:10 loss 1.2048 (1.1035) acc 100.0000 (99.6875) lr 0.026000 -epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1027 (1.1107) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.352, TIME@all 0.312 -epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.1960 (1.0974) acc 100.0000 (100.0000) lr 0.026000 -epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:07 loss 1.1064 (1.0968) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.154, TIME@all 0.313 -epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.0960 (1.0921) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1279 (1.1049) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.271, TIME@all 0.312 -epoch: [162/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:49:11 loss 1.1384 (1.1116) acc 96.8750 (99.3750) lr 0.026000 -epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1473 (1.1108) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.281, TIME@all 0.312 -epoch: [162/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:49:11 loss 1.1183 (1.0939) acc 100.0000 (100.0000) lr 0.026000 -epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.1246 (1.1082) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.194, TIME@all 0.313 -epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:49:11 loss 1.1065 (1.0953) acc 100.0000 (99.8438) lr 0.026000 -epoch: [162/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:49:06 loss 1.0991 (1.1007) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.198, TIME@all 0.313 -epoch: [162/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:49:11 loss 1.1042 (1.0960) acc 100.0000 (100.0000) lr 0.026000 -epoch: [162/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:49:06 loss 1.1819 (1.1070) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.185, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:48:56 loss 1.1592 (1.1137) acc 100.0000 (99.6875) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:48:50 loss 1.1122 (1.1193) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.046, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:48:56 loss 1.0961 (1.0920) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:48:50 loss 1.1262 (1.1021) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.086, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1075 (1.0905) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1375 (1.1038) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.974, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1247 (1.1102) acc 100.0000 (99.5312) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0967 (1.1128) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 818.971, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.0978 (1.0943) acc 100.0000 (100.0000) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1109 (1.1132) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.015, TIME@all 0.313 -epoch: [163/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1438 (1.1087) acc 100.0000 (99.3750) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.1043 (1.1147) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 818.984, TIME@all 0.313 -epoch: [163/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.1187 (1.0998) acc 100.0000 (99.6875) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0872 (1.1045) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.026, TIME@all 0.313 -epoch: [163/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:48:56 loss 1.0996 (1.0999) acc 100.0000 (99.6875) lr 0.026000 -epoch: [163/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:48:50 loss 1.0865 (1.1033) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.963, TIME@all 0.313 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:48:33 loss 1.1142 (1.0848) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.0996 (1.0964) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.958, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1318 (1.0932) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.2026 (1.1110) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 820.773, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:48:33 loss 1.2138 (1.1032) acc 96.8750 (99.6875) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.0983 (1.1075) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.871, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:48:34 loss 1.1922 (1.0897) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:48:27 loss 1.0885 (1.1084) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 820.778, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1330 (1.0920) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1383 (1.1084) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.836, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:33 loss 1.1036 (1.0885) acc 100.0000 (100.0000) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1250 (1.1035) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.876, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:33 loss 1.1117 (1.0925) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1515 (1.1060) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.887, TIME@all 0.312 -epoch: [164/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:48:34 loss 1.1593 (1.1019) acc 100.0000 (99.8438) lr 0.026000 -epoch: [164/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:48:27 loss 1.1226 (1.1087) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.797, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.2141 (1.0954) acc 100.0000 (99.8438) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:48:13 loss 1.1854 (1.1004) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 820.488, TIME@all 0.312 -epoch: [165/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:48:17 loss 1.2252 (1.1000) acc 96.8750 (99.8438) lr 0.026000 -epoch: [165/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:48:13 loss 1.0970 (1.1021) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.530, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.2637 (1.1052) acc 100.0000 (99.6875) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:14 loss 1.1583 (1.1130) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 820.361, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.2306 (1.0983) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:13 loss 1.1546 (1.1030) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.411, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.1858 (1.0975) acc 96.8750 (99.6875) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:48:13 loss 1.1369 (1.1086) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.367, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.1564 (1.1072) acc 96.8750 (99.6875) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:48:13 loss 1.1066 (1.1085) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.429, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:48:18 loss 1.2006 (1.0953) acc 100.0000 (99.8438) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:48:13 loss 1.1920 (1.1070) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.441, TIME@all 0.312 -epoch: [165/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:48:18 loss 1.1919 (1.0915) acc 100.0000 (100.0000) lr 0.026000 -epoch: [165/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:48:13 loss 1.1109 (1.0990) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 820.426, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:48:12 loss 1.1346 (1.0974) acc 100.0000 (99.8438) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1516 (1.1135) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.617, TIME@all 0.312 -epoch: [166/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:48:11 loss 1.1108 (1.0892) acc 100.0000 (100.0000) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:47:58 loss 1.0916 (1.1113) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.698, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:48:12 loss 1.1281 (1.0945) acc 96.8750 (99.6875) lr 0.026000 -epoch: [166/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.0862 (1.1044) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.560, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:48:12 loss 1.1439 (1.0873) acc 100.0000 (100.0000) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:59 loss 1.0924 (1.1007) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.506, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:48:12 loss 1.1028 (1.0991) acc 100.0000 (99.6875) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1913 (1.1073) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 820.598, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:48:12 loss 1.1840 (1.1064) acc 96.8750 (99.5312) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1677 (1.1119) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 820.544, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:48:12 loss 1.1997 (1.0962) acc 96.8750 (99.6875) lr 0.026000 -epoch: [166/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:47:58 loss 1.0705 (1.1105) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 820.542, TIME@all 0.312 -epoch: [166/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:48:12 loss 1.1511 (1.0983) acc 100.0000 (100.0000) lr 0.026000 -epoch: [166/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:47:58 loss 1.1043 (1.1112) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.557, TIME@all 0.312 -epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:47:49 loss 1.0973 (1.0916) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0940 (1.1062) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.604, TIME@all 0.312 -epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0856 (1.0875) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:47:44 loss 1.0758 (1.0963) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.681, TIME@all 0.312 -epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.1391 (1.0911) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1359 (1.1043) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.439, TIME@all 0.312 -epoch: [167/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.1176 (1.0918) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1370 (1.1017) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.496, TIME@all 0.312 -epoch: [167/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:47:49 loss 1.0893 (1.0890) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0972 (1.0984) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.587, TIME@all 0.312 -epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:47:49 loss 1.0700 (1.0930) acc 100.0000 (99.8438) lr 0.026000 -epoch: [167/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:47:45 loss 1.1173 (1.1046) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.534, TIME@all 0.312 -epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0865 (1.0916) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.317 (0.313) data 0.000 (0.007) eta 0:47:45 loss 1.0852 (1.0998) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.532, TIME@all 0.312 -epoch: [167/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:47:49 loss 1.0713 (1.0870) acc 100.0000 (100.0000) lr 0.026000 -epoch: [167/350][40/50] time 0.316 (0.313) data 0.001 (0.007) eta 0:47:45 loss 1.0869 (1.1023) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.583, TIME@all 0.312 -epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1027 (1.0957) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0798 (1.1076) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 818.812, TIME@all 0.313 -epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.014) eta 0:47:46 loss 1.1694 (1.0993) acc 100.0000 (100.0000) lr 0.026000 -epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.2112 (1.1152) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 818.869, TIME@all 0.313 -epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.0752 (1.0906) acc 100.0000 (100.0000) lr 0.026000 -epoch: [168/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:47:35 loss 1.0830 (1.1016) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.720, TIME@all 0.313 -epoch: [168/350][20/50] time 0.313 (0.314) data 0.001 (0.014) eta 0:47:47 loss 1.1385 (1.0855) acc 96.8750 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0720 (1.0980) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.751, TIME@all 0.313 -epoch: [168/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1324 (1.0977) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.1014 (1.0996) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.731, TIME@all 0.313 -epoch: [168/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:46 loss 1.1331 (1.0927) acc 100.0000 (99.6875) lr 0.026000 -epoch: [168/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0698 (1.0966) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.778, TIME@all 0.313 -epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.1511 (1.0997) acc 100.0000 (99.5312) lr 0.026000 -epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.1283 (1.1067) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 818.810, TIME@all 0.313 -epoch: [168/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:47 loss 1.0797 (1.0908) acc 100.0000 (99.8438) lr 0.026000 -epoch: [168/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:47:34 loss 1.0974 (1.1011) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.819, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0705 (1.0833) acc 100.0000 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.1256 (1.0956) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.396, TIME@all 0.313 -epoch: [169/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:47:27 loss 1.0937 (1.0842) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0926 (1.0932) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 818.457, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:47:27 loss 1.1603 (1.0888) acc 96.8750 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:19 loss 1.1415 (1.0995) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 818.521, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:28 loss 1.0879 (1.0807) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:47:20 loss 1.1288 (1.0971) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.291, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0992 (1.0893) acc 100.0000 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0722 (1.1010) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.395, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:27 loss 1.0687 (1.0865) acc 100.0000 (99.8438) lr 0.026000 -epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.0996 (1.0990) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.358, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:47:28 loss 1.1063 (1.0836) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:47:20 loss 1.1410 (1.1017) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 818.347, TIME@all 0.313 -epoch: [169/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:47:28 loss 1.0964 (1.0828) acc 100.0000 (100.0000) lr 0.026000 -epoch: [169/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:47:20 loss 1.0907 (1.0918) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 818.321, TIME@all 0.313 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:02 loss 1.0937 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:56 loss 1.0953 (1.0959) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.663, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:02 loss 1.1135 (1.0802) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:57 loss 1.1537 (1.0966) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.561, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:47:03 loss 1.1011 (1.0822) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:46:57 loss 1.0950 (1.1013) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.500, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:47:03 loss 1.1051 (1.0851) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:46:57 loss 1.0788 (1.0976) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.477, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1104 (1.0899) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:46:57 loss 1.1646 (1.1015) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.558, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1073 (1.0843) acc 100.0000 (100.0000) lr 0.026000 -epoch: [170/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 0:46:57 loss 1.1065 (1.1019) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.523, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:47:03 loss 1.1341 (1.0893) acc 96.8750 (99.6875) lr 0.026000 -epoch: [170/350][40/50] time 0.318 (0.313) data 0.001 (0.006) eta 0:46:57 loss 1.1000 (1.0996) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.529, TIME@all 0.312 -epoch: [170/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:47:03 loss 1.1169 (1.0876) acc 100.0000 (99.8438) lr 0.026000 -epoch: [170/350][40/50] time 0.317 (0.313) data 0.001 (0.007) eta 0:46:57 loss 1.0914 (1.0996) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.596, TIME@all 0.312 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1229 (1.0882) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0995 (1.1024) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.332, TIME@all 0.313 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:00 loss 1.1308 (1.0905) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0990 (1.1070) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.394, TIME@all 0.313 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.0923 (1.0846) acc 100.0000 (100.0000) lr 0.026000 -epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0714 (1.1027) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.230, TIME@all 0.313 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.0838 (1.0913) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.1146 (1.1066) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.264, TIME@all 0.313 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1385 (1.0885) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0776 (1.1052) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 818.295, TIME@all 0.313 -epoch: [171/350][20/50] time 0.313 (0.314) data 0.001 (0.012) eta 0:47:01 loss 1.1733 (1.0877) acc 100.0000 (100.0000) lr 0.026000 -epoch: [171/350][40/50] time 0.309 (0.313) data 0.001 (0.006) eta 0:46:47 loss 1.0868 (1.1044) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 818.315, TIME@all 0.313 -epoch: [171/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:47:01 loss 1.1740 (1.0959) acc 100.0000 (99.8438) lr 0.026000 -epoch: [171/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:46:47 loss 1.0869 (1.1154) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.283, TIME@all 0.313 -epoch: [171/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:47:01 loss 1.1515 (1.0951) acc 96.8750 (99.6875) lr 0.026000 -epoch: [171/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:46:47 loss 1.0845 (1.1131) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 818.279, TIME@all 0.313 -epoch: [172/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0968 (1.0907) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:46:24 loss 1.0998 (1.0977) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 820.568, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1791 (1.0930) acc 96.8750 (99.6875) lr 0.026000 -epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1489 (1.0997) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.437, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0867 (1.0841) acc 100.0000 (99.8438) lr 0.026000 -epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:46:24 loss 1.0962 (1.0954) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.486, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1160 (1.0914) acc 100.0000 (99.8438) lr 0.026000 -epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1288 (1.1008) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.463, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:29 loss 1.1294 (1.0913) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1186 (1.1027) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.333, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.0961 (1.0935) acc 100.0000 (99.6875) lr 0.026000 -epoch: [172/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1098 (1.0985) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.414, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:46:28 loss 1.0908 (1.0879) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:46:24 loss 1.1095 (1.0998) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.449, TIME@all 0.312 -epoch: [172/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:46:28 loss 1.1324 (1.0869) acc 100.0000 (100.0000) lr 0.026000 -epoch: [172/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:46:24 loss 1.1002 (1.1024) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.504, TIME@all 0.312 -epoch: [173/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:46:13 loss 1.0876 (1.0781) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1310 (1.0955) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.870, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1129 (1.0868) acc 96.8750 (99.5312) lr 0.026000 -epoch: [173/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1593 (1.0990) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 818.823, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1344 (1.0817) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:46:13 loss 1.2195 (1.1031) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 818.736, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1469 (1.0876) acc 100.0000 (99.8438) lr 0.026000 -epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1992 (1.1056) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 818.710, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:46:13 loss 1.1506 (1.0815) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:46:13 loss 1.1593 (1.0989) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 818.790, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1098 (1.0833) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1231 (1.1005) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.724, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.2137 (1.0838) acc 96.8750 (99.8438) lr 0.026000 -epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.1481 (1.0952) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 818.770, TIME@all 0.313 -epoch: [173/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:46:13 loss 1.1142 (1.0846) acc 100.0000 (100.0000) lr 0.026000 -epoch: [173/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:46:13 loss 1.0898 (1.0961) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.780, TIME@all 0.313 -epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:00 loss 1.0744 (1.0829) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:52 loss 1.2057 (1.1027) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 820.797, TIME@all 0.312 -epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0790 (1.0850) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:45:53 loss 1.1632 (1.0992) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.602, TIME@all 0.312 -epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:46:00 loss 1.0787 (1.0848) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:52 loss 1.1333 (1.0954) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 820.725, TIME@all 0.312 -epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0685 (1.0851) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.1209 (1.0969) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.688, TIME@all 0.312 -epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:46:00 loss 1.0823 (1.0853) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.1730 (1.1015) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 820.646, TIME@all 0.312 -epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.0851 (1.0780) acc 100.0000 (99.8438) lr 0.026000 -epoch: [174/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:45:52 loss 1.0951 (1.0916) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.671, TIME@all 0.312 -epoch: [174/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:46:01 loss 1.1121 (1.0833) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:45:53 loss 1.0994 (1.0961) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.634, TIME@all 0.312 -epoch: [174/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:46:01 loss 1.1133 (1.0836) acc 100.0000 (100.0000) lr 0.026000 -epoch: [174/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:45:52 loss 1.0871 (1.0953) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.685, TIME@all 0.312 -epoch: [175/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0872 (1.0789) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0642 (1.0897) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.142, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0821 (1.0812) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0927 (1.0920) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 816.163, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:45:59 loss 1.0971 (1.0806) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:54 loss 1.0799 (1.0912) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.051, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:45:59 loss 1.0854 (1.0818) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.315) data 0.000 (0.006) eta 0:45:55 loss 1.0846 (1.0895) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 816.003, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0808 (1.0826) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0984 (1.0857) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 816.081, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0631 (1.0815) acc 100.0000 (99.6875) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0813 (1.0941) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 816.052, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0745 (1.0800) acc 100.0000 (99.8438) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:54 loss 1.0762 (1.0890) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 816.075, TIME@all 0.314 -epoch: [175/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:45:59 loss 1.0924 (1.0766) acc 100.0000 (100.0000) lr 0.026000 -epoch: [175/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:54 loss 1.0700 (1.0923) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.057, TIME@all 0.314 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.014) eta 0:45:38 loss 1.0942 (1.0837) acc 100.0000 (99.5312) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1388 (1.0942) acc 100.0000 (99.6875) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.733, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1086 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0991 (1.0995) acc 100.0000 (99.6875) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.776, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0848 (1.0798) acc 100.0000 (99.8438) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.2606 (1.0932) acc 96.8750 (99.7656) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.616, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1308 (1.0925) acc 96.8750 (99.6875) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:45:32 loss 1.0989 (1.0997) acc 100.0000 (99.6094) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.634, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1003 (1.0937) acc 100.0000 (99.5312) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1113 (1.0985) acc 100.0000 (99.7656) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.673, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.1042 (1.0808) acc 100.0000 (100.0000) lr 0.026000 -epoch: [176/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0915 (1.0976) acc 100.0000 (99.8438) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.753, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0926 (1.0853) acc 100.0000 (99.8438) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.0981 (1.0979) acc 100.0000 (99.8438) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.657, TIME@all 0.313 -epoch: [176/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:45:39 loss 1.0752 (1.0847) acc 100.0000 (100.0000) lr 0.026000 -epoch: [176/350][40/50] time 0.311 (0.314) data 0.000 (0.007) eta 0:45:32 loss 1.1506 (1.0980) acc 100.0000 (99.9219) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 818.718, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1169 (1.0868) acc 100.0000 (99.6875) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:12 loss 1.0959 (1.1008) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.280, TIME@all 0.312 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.0809 (1.0811) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:13 loss 1.1460 (1.0998) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.122, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:17 loss 1.0847 (1.0983) acc 100.0000 (99.3750) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:45:12 loss 1.1162 (1.1003) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.298, TIME@all 0.312 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:45:18 loss 1.0858 (1.0868) acc 100.0000 (99.8438) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:45:13 loss 1.0993 (1.0981) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.130, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1589 (1.0851) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:45:12 loss 1.1792 (1.0994) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 819.198, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.2168 (1.0856) acc 96.8750 (99.6875) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:45:13 loss 1.1056 (1.1010) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.124, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.1494 (1.0851) acc 100.0000 (99.8438) lr 0.026000 -epoch: [177/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:45:13 loss 1.1320 (1.0928) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.172, TIME@all 0.313 -epoch: [177/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:45:18 loss 1.0915 (1.0764) acc 100.0000 (100.0000) lr 0.026000 -epoch: [177/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:45:13 loss 1.0856 (1.0897) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.169, TIME@all 0.313 -epoch: [178/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:45:00 loss 1.1846 (1.0917) acc 100.0000 (99.6875) lr 0.026000 -epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0805 (1.1035) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.257, TIME@all 0.312 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:44:59 loss 1.1686 (1.0946) acc 100.0000 (99.5312) lr 0.026000 -epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1024 (1.1070) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.312, TIME@all 0.312 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0808 (1.0880) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0874 (1.0989) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.114, TIME@all 0.313 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0755 (1.0885) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1078 (1.1000) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.157, TIME@all 0.313 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.0919 (1.0843) acc 100.0000 (100.0000) lr 0.026000 -epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0657 (1.0916) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.226, TIME@all 0.312 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1188 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0871 (1.0958) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.232, TIME@all 0.312 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1162 (1.0908) acc 100.0000 (99.6875) lr 0.026000 -epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.0921 (1.0961) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.216, TIME@all 0.312 -epoch: [178/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:44:59 loss 1.1376 (1.0833) acc 100.0000 (99.8438) lr 0.026000 -epoch: [178/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:44:54 loss 1.1025 (1.1024) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.220, TIME@all 0.312 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.014) eta 0:44:59 loss 1.1091 (1.0826) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.0947 (1.0883) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 815.478, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.0868 (1.0776) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1072 (1.0961) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 815.406, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:45:00 loss 1.0849 (1.0859) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1382 (1.0966) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 815.301, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1089 (1.0862) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.316 (0.315) data 0.001 (0.007) eta 0:44:57 loss 1.1127 (1.0929) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 815.352, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1143 (1.0760) acc 96.8750 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.2115 (1.0898) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 815.336, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.0968 (1.0783) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.2125 (1.0970) acc 93.7500 (99.8438) lr 0.026000 -FPS@all 815.393, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:44:59 loss 1.1363 (1.0831) acc 100.0000 (99.8438) lr 0.026000 -epoch: [179/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:44:56 loss 1.0930 (1.0944) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 815.390, TIME@all 0.314 -epoch: [179/350][20/50] time 0.316 (0.315) data 0.000 (0.014) eta 0:44:59 loss 1.0902 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [179/350][40/50] time 0.316 (0.315) data 0.000 (0.007) eta 0:44:57 loss 1.1532 (1.0908) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 815.375, TIME@all 0.314 -epoch: [180/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:44:25 loss 1.1041 (1.0857) acc 100.0000 (100.0000) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:19 loss 1.0855 (1.0986) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.558, TIME@all 0.312 -epoch: [180/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:44:25 loss 1.2310 (1.0939) acc 96.8750 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0602 (1.1018) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.433, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:26 loss 1.2113 (1.0956) acc 96.8750 (99.5312) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0845 (1.0998) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.291, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:44:25 loss 1.1287 (1.0869) acc 100.0000 (99.8438) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.1811 (1.0990) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 820.345, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.0828 (1.0895) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.1393 (1.1009) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.395, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.2384 (1.0958) acc 96.8750 (99.8438) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0908 (1.1009) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.388, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:44:25 loss 1.0874 (1.0851) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0952 (1.0984) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.387, TIME@all 0.312 -epoch: [180/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:44:25 loss 1.1219 (1.0877) acc 100.0000 (99.6875) lr 0.026000 -epoch: [180/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:44:20 loss 1.0853 (1.0913) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.353, TIME@all 0.312 -epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.0907 (1.0975) acc 100.0000 (99.6875) lr 0.026000 -epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1287 (1.0981) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.391, TIME@all 0.313 -epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:44:23 loss 1.1234 (1.0879) acc 100.0000 (100.0000) lr 0.026000 -epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1072 (1.0991) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.404, TIME@all 0.313 -epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:44:24 loss 1.0822 (1.0906) acc 100.0000 (99.6875) lr 0.026000 -epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:44:17 loss 1.0925 (1.0969) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.235, TIME@all 0.313 -epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:44:24 loss 1.0653 (1.0933) acc 100.0000 (99.6875) lr 0.026000 -epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:17 loss 1.0901 (1.0946) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 817.284, TIME@all 0.313 -epoch: [181/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.0945 (1.0923) acc 100.0000 (99.8438) lr 0.026000 -epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1381 (1.1011) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 817.360, TIME@all 0.313 -epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:24 loss 1.0801 (1.0940) acc 100.0000 (99.5312) lr 0.026000 -epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:17 loss 1.1612 (1.0940) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 817.308, TIME@all 0.313 -epoch: [181/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:44:23 loss 1.1109 (1.0932) acc 100.0000 (99.8438) lr 0.026000 -epoch: [181/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.1168 (1.0966) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 817.348, TIME@all 0.313 -epoch: [181/350][20/50] time 0.316 (0.314) data 0.001 (0.013) eta 0:44:24 loss 1.0730 (1.0886) acc 100.0000 (100.0000) lr 0.026000 -epoch: [181/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:44:16 loss 1.0738 (1.0981) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 817.319, TIME@all 0.313 -epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:44:01 loss 1.1048 (1.0785) acc 100.0000 (99.6875) lr 0.026000 -epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.1112 (1.0877) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.036, TIME@all 0.313 -epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:44:00 loss 1.0998 (1.0780) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.1174 (1.0841) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.083, TIME@all 0.313 -epoch: [182/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.1045 (1.0744) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1242 (1.0860) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 818.974, TIME@all 0.313 -epoch: [182/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.0811 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1171 (1.0840) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.954, TIME@all 0.313 -epoch: [182/350][20/50] time 0.319 (0.313) data 0.001 (0.012) eta 0:44:01 loss 1.0858 (1.0793) acc 100.0000 (100.0000) lr 0.026000 -epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1780 (1.0893) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.042, TIME@all 0.313 -epoch: [182/350][20/50] time 0.318 (0.313) data 0.001 (0.013) eta 0:44:01 loss 1.0974 (1.0836) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.0862 (1.0894) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.933, TIME@all 0.313 -epoch: [182/350][20/50] time 0.319 (0.313) data 0.000 (0.012) eta 0:44:01 loss 1.0696 (1.0736) acc 100.0000 (99.8438) lr 0.026000 -epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:43:55 loss 1.1595 (1.0895) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.037, TIME@all 0.313 -epoch: [182/350][20/50] time 0.319 (0.313) data 0.001 (0.013) eta 0:44:00 loss 1.2617 (1.0860) acc 96.8750 (99.6875) lr 0.026000 -epoch: [182/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:43:55 loss 1.0929 (1.0882) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.049, TIME@all 0.313 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:34 loss 1.0841 (1.0883) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:43:31 loss 1.1061 (1.0936) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.674, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:34 loss 1.0882 (1.0909) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:43:31 loss 1.1100 (1.0982) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.722, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1480 (1.0976) acc 100.0000 (99.5312) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:32 loss 1.0966 (1.0996) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.528, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.0924 (1.0894) acc 100.0000 (99.8438) lr 0.026000 -epoch: [183/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.2214 (1.1010) acc 93.7500 (99.6875) lr 0.026000 -FPS@all 819.606, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1171 (1.0787) acc 100.0000 (100.0000) lr 0.026000 -epoch: [183/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.0845 (1.0866) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.578, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:43:35 loss 1.1461 (1.0857) acc 96.8750 (99.8438) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:43:31 loss 1.1166 (1.0934) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.586, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1418 (1.1041) acc 100.0000 (99.6875) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.1194 (1.1032) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.633, TIME@all 0.312 -epoch: [183/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:43:35 loss 1.1208 (1.0948) acc 100.0000 (99.6875) lr 0.026000 -epoch: [183/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:43:31 loss 1.0839 (1.1015) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.650, TIME@all 0.312 -epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:28 loss 1.0964 (1.0773) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.316 (0.313) data 0.000 (0.007) eta 0:43:20 loss 1.0848 (1.0877) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.055, TIME@all 0.312 -epoch: [184/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:43:29 loss 1.0755 (1.0793) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:43:21 loss 1.1577 (1.0928) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.898, TIME@all 0.312 -epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0704 (1.0772) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0929 (1.0895) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.948, TIME@all 0.312 -epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:28 loss 1.0958 (1.0834) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:20 loss 1.1336 (1.0943) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.975, TIME@all 0.312 -epoch: [184/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0695 (1.0778) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.1108 (1.0938) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.943, TIME@all 0.312 -epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0921 (1.0801) acc 100.0000 (100.0000) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.001 (0.007) eta 0:43:21 loss 1.1255 (1.0896) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.856, TIME@all 0.312 -epoch: [184/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0665 (1.0783) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0920 (1.0896) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.942, TIME@all 0.312 -epoch: [184/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:29 loss 1.0706 (1.0816) acc 100.0000 (99.8438) lr 0.026000 -epoch: [184/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:43:21 loss 1.0795 (1.0876) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.946, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.014) eta 0:43:09 loss 1.1366 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0602 (1.0937) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.430, TIME@all 0.312 -epoch: [185/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1594 (1.0780) acc 96.8750 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0764 (1.0901) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.276, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:43:09 loss 1.1164 (1.0807) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:43:02 loss 1.0860 (1.0936) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.278, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.0793 (1.0821) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0794 (1.0978) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 820.365, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1308 (1.0862) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0747 (1.1008) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.292, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.0801 (1.0837) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:43:02 loss 1.0686 (1.0948) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.279, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:10 loss 1.0711 (1.0796) acc 100.0000 (99.8438) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:43:02 loss 1.0605 (1.0915) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.277, TIME@all 0.312 -epoch: [185/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:43:09 loss 1.1527 (1.0813) acc 100.0000 (100.0000) lr 0.026000 -epoch: [185/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:43:02 loss 1.0653 (1.0933) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.283, TIME@all 0.312 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:56 loss 1.1150 (1.0760) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:42:50 loss 1.0780 (1.0868) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.958, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.1010 (1.0841) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0701 (1.0926) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 818.829, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0937 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0705 (1.0839) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.823, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0944 (1.0742) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0566 (1.0837) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.849, TIME@all 0.313 -epoch: [186/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0973 (1.0732) acc 100.0000 (99.8438) lr 0.026000 -epoch: [186/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0660 (1.0851) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.845, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:42:57 loss 1.0899 (1.0726) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:42:50 loss 1.0827 (1.0802) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.870, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.1177 (1.0755) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0818 (1.0832) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.916, TIME@all 0.313 -epoch: [186/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:42:57 loss 1.0888 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [186/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:42:50 loss 1.0815 (1.0873) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.894, TIME@all 0.313 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:38 loss 1.1360 (1.0895) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.309 (0.313) data 0.001 (0.007) eta 0:42:31 loss 1.2239 (1.0915) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.929, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.312) data 0.000 (0.014) eta 0:42:34 loss 1.0961 (1.0832) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:30 loss 1.1490 (1.0943) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.329, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0764 (1.0802) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1886 (1.0893) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 819.769, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.1471 (1.0824) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.1412 (1.0920) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.750, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0988 (1.0844) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1115 (1.0934) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.758, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:42:39 loss 1.1327 (1.0786) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:42:32 loss 1.1840 (1.0922) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.810, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:39 loss 1.0997 (1.0835) acc 100.0000 (99.8438) lr 0.026000 -epoch: [187/350][40/50] time 0.309 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.0744 (1.0900) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.827, TIME@all 0.312 -epoch: [187/350][20/50] time 0.308 (0.313) data 0.000 (0.014) eta 0:42:39 loss 1.1215 (1.0898) acc 100.0000 (100.0000) lr 0.026000 -epoch: [187/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:42:32 loss 1.1662 (1.0909) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.799, TIME@all 0.312 -epoch: [188/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.1162 (1.0744) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.0871 (1.0904) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 821.125, TIME@all 0.312 -epoch: [188/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.0874 (1.0717) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:12 loss 1.0756 (1.0835) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.168, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:19 loss 1.0838 (1.0745) acc 100.0000 (99.8438) lr 0.026000 -epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.0616 (1.0912) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 821.033, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:18 loss 1.0719 (1.0737) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0901 (1.0862) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 821.120, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:19 loss 1.0698 (1.0732) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0734 (1.0854) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.063, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:42:19 loss 1.1190 (1.0783) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:42:13 loss 1.0709 (1.0858) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.062, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:18 loss 1.0813 (1.0835) acc 100.0000 (99.8438) lr 0.026000 -epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.1038 (1.0977) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 821.113, TIME@all 0.312 -epoch: [188/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:42:18 loss 1.1076 (1.0761) acc 100.0000 (100.0000) lr 0.026000 -epoch: [188/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:42:13 loss 1.1010 (1.0855) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.113, TIME@all 0.312 -epoch: [189/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1836 (1.0847) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:42:02 loss 1.0671 (1.0933) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.084, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:42:11 loss 1.1868 (1.0739) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0645 (1.0922) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.969, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.1124 (1.0800) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0681 (1.0967) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.964, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1013 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:42:03 loss 1.0829 (1.0925) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.031, TIME@all 0.313 -epoch: [189/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:42:11 loss 1.1601 (1.0769) acc 100.0000 (100.0000) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.1035 (1.0878) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.990, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.0813 (1.0781) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0680 (1.0907) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.967, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.001 (0.013) eta 0:42:11 loss 1.0770 (1.0785) acc 100.0000 (99.6875) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:42:03 loss 1.0708 (1.0885) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.971, TIME@all 0.313 -epoch: [189/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:42:11 loss 1.0991 (1.0824) acc 100.0000 (99.8438) lr 0.026000 -epoch: [189/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:42:03 loss 1.0802 (1.0936) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.013, TIME@all 0.313 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0831 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.0850 (1.0822) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.411, TIME@all 0.312 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0711 (1.0693) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1006 (1.0814) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 820.464, TIME@all 0.312 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:41:54 loss 1.0891 (1.0714) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:41:43 loss 1.1341 (1.0880) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.290, TIME@all 0.312 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0628 (1.0748) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1630 (1.0929) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.319, TIME@all 0.312 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:41:54 loss 1.0920 (1.0747) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1240 (1.0889) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.422, TIME@all 0.312 -epoch: [190/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0547 (1.0809) acc 100.0000 (99.6875) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.1035 (1.0904) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.368, TIME@all 0.312 -epoch: [190/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:54 loss 1.0931 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:42 loss 1.0784 (1.0835) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.367, TIME@all 0.312 -epoch: [190/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:41:54 loss 1.0701 (1.0773) acc 100.0000 (99.8438) lr 0.026000 -epoch: [190/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:41:42 loss 1.0862 (1.0882) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.364, TIME@all 0.312 -epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:41:42 loss 1.0992 (1.0748) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:41:27 loss 1.1787 (1.0911) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.216, TIME@all 0.312 -epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1026 (1.0769) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1635 (1.0843) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.146, TIME@all 0.313 -epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.0933 (1.0764) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1343 (1.0833) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.111, TIME@all 0.313 -epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1142 (1.0794) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1196 (1.0903) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.084, TIME@all 0.313 -epoch: [191/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.0963 (1.0777) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:41:27 loss 1.1312 (1.0915) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.136, TIME@all 0.313 -epoch: [191/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:41:43 loss 1.0751 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:41:27 loss 1.2022 (1.0866) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 819.130, TIME@all 0.313 -epoch: [191/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:41:43 loss 1.1993 (1.0860) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1359 (1.0894) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.099, TIME@all 0.313 -epoch: [191/350][20/50] time 0.316 (0.314) data 0.001 (0.012) eta 0:41:43 loss 1.0862 (1.0737) acc 100.0000 (99.8438) lr 0.026000 -epoch: [191/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:41:27 loss 1.1705 (1.0892) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.150, TIME@all 0.313 -epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:41:24 loss 1.0751 (1.0709) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:41:14 loss 1.1225 (1.0899) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.540, TIME@all 0.312 -epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1043 (1.0691) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.2503 (1.0904) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.425, TIME@all 0.312 -epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0770 (1.0791) acc 100.0000 (99.6875) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.0817 (1.0858) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.445, TIME@all 0.312 -epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0724 (1.0683) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:14 loss 1.0960 (1.0848) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.462, TIME@all 0.312 -epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:41:25 loss 1.0802 (1.0737) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:41:15 loss 1.1527 (1.0931) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.442, TIME@all 0.312 -epoch: [192/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1003 (1.0717) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1241 (1.0822) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.478, TIME@all 0.312 -epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.0635 (1.0733) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1435 (1.0925) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.490, TIME@all 0.312 -epoch: [192/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:41:25 loss 1.1038 (1.0712) acc 100.0000 (100.0000) lr 0.026000 -epoch: [192/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:41:15 loss 1.1047 (1.0832) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.476, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.0596 (1.0721) acc 100.0000 (99.8438) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.0818 (1.0817) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.336, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.0836 (1.0730) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.1505 (1.0855) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 820.245, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:01 loss 1.0662 (1.0786) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1196 (1.0865) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.167, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0692 (1.0698) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1867 (1.0818) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 820.222, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.1091 (1.0857) acc 100.0000 (99.8438) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.1121 (1.0880) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.238, TIME@all 0.312 -epoch: [193/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:41:00 loss 1.1040 (1.0791) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:40:57 loss 1.0852 (1.0796) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 820.226, TIME@all 0.312 -epoch: [193/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0816 (1.0791) acc 100.0000 (99.6875) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1292 (1.0857) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.234, TIME@all 0.312 -epoch: [193/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:41:00 loss 1.0940 (1.0797) acc 100.0000 (100.0000) lr 0.026000 -epoch: [193/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:40:57 loss 1.1539 (1.0844) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 820.225, TIME@all 0.312 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:58 loss 1.1249 (1.0852) acc 100.0000 (99.6875) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0630 (1.0917) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 814.875, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.1177 (1.0774) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0804 (1.0889) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.837, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.011) eta 0:40:59 loss 1.0782 (1.0758) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0755 (1.0817) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.765, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0740 (1.0711) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.1151 (1.0827) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.683, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.1320 (1.0824) acc 100.0000 (99.6875) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0933 (1.0859) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 814.782, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:58 loss 1.0941 (1.0737) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0646 (1.0896) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.864, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0655 (1.0712) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.1715 (1.0873) acc 96.8750 (99.9219) lr 0.026000 -FPS@all 814.779, TIME@all 0.314 -epoch: [194/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:40:59 loss 1.0687 (1.0744) acc 100.0000 (100.0000) lr 0.026000 -epoch: [194/350][40/50] time 0.312 (0.315) data 0.000 (0.006) eta 0:41:02 loss 1.0723 (1.0818) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 814.804, TIME@all 0.314 -epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1256 (1.0767) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1168 (1.0888) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.086, TIME@all 0.313 -epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:35 loss 1.1625 (1.0807) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1465 (1.0937) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 817.931, TIME@all 0.313 -epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1195 (1.0751) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1115 (1.0882) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.017, TIME@all 0.313 -epoch: [195/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.0772 (1.0763) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1221 (1.0889) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 817.979, TIME@all 0.313 -epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.0884 (1.0734) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.0797 (1.0807) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 817.982, TIME@all 0.313 -epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1106 (1.0741) acc 100.0000 (100.0000) lr 0.026000 -epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:33 loss 1.1217 (1.0848) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.000, TIME@all 0.313 -epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:40:35 loss 1.0772 (1.0741) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:33 loss 1.0877 (1.0867) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 817.957, TIME@all 0.313 -epoch: [195/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:40:34 loss 1.1295 (1.0838) acc 100.0000 (99.8438) lr 0.026000 -epoch: [195/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:33 loss 1.1331 (1.0981) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 817.997, TIME@all 0.313 -epoch: [196/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.0739 (1.0758) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:40:06 loss 1.1075 (1.0827) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.194, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.0734 (1.0755) acc 100.0000 (99.8438) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 0:40:07 loss 1.1001 (1.0881) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 821.102, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.1167 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.001 (0.006) eta 0:40:07 loss 1.0582 (1.0881) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 821.140, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.0679 (1.0814) acc 100.0000 (99.6875) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.1522 (1.0879) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 821.072, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.1188 (1.0793) acc 100.0000 (99.8438) lr 0.026000 -epoch: [196/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:40:06 loss 1.1090 (1.0856) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 821.160, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.1483 (1.0859) acc 100.0000 (99.6875) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:40:07 loss 1.0743 (1.0942) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 821.085, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:40:10 loss 1.1027 (1.0743) acc 100.0000 (100.0000) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.1133 (1.0876) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 821.096, TIME@all 0.312 -epoch: [196/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:40:10 loss 1.0927 (1.0833) acc 100.0000 (99.6875) lr 0.026000 -epoch: [196/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:40:07 loss 1.0887 (1.0881) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 821.106, TIME@all 0.312 -epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0733 (1.0738) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:40:08 loss 1.0862 (1.0890) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.732, TIME@all 0.314 -epoch: [197/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0572 (1.0715) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1123 (1.0845) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.643, TIME@all 0.314 -epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0634 (1.0794) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0982 (1.0911) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.596, TIME@all 0.314 -epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:16 loss 1.0723 (1.0728) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:40:08 loss 1.1479 (1.0880) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 814.756, TIME@all 0.314 -epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0657 (1.0689) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0859 (1.0956) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 814.607, TIME@all 0.314 -epoch: [197/350][20/50] time 0.319 (0.315) data 0.001 (0.013) eta 0:40:18 loss 1.0705 (1.0777) acc 100.0000 (100.0000) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1273 (1.0895) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.617, TIME@all 0.314 -epoch: [197/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:40:17 loss 1.0664 (1.0750) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.317 (0.315) data 0.000 (0.007) eta 0:40:09 loss 1.1277 (1.0911) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 814.654, TIME@all 0.314 -epoch: [197/350][20/50] time 0.319 (0.315) data 0.000 (0.012) eta 0:40:17 loss 1.0891 (1.0829) acc 100.0000 (99.8438) lr 0.026000 -epoch: [197/350][40/50] time 0.316 (0.315) data 0.000 (0.006) eta 0:40:09 loss 1.0942 (1.0951) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 814.676, TIME@all 0.314 -epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.013) eta 0:42:08 loss 1.0687 (1.0764) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.315 (0.321) data 0.000 (0.007) eta 0:40:45 loss 1.0764 (1.0888) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 802.518, TIME@all 0.319 -epoch: [198/350][20/50] time 0.329 (0.331) data 0.001 (0.014) eta 0:42:06 loss 1.1264 (1.0802) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.315 (0.321) data 0.001 (0.007) eta 0:40:44 loss 1.0906 (1.0909) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 802.769, TIME@all 0.319 -epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.012) eta 0:42:09 loss 1.0881 (1.0721) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.006) eta 0:40:45 loss 1.0808 (1.0801) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 802.338, TIME@all 0.319 -epoch: [198/350][20/50] time 0.329 (0.331) data 0.001 (0.013) eta 0:42:06 loss 1.0723 (1.0688) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:44 loss 1.0827 (1.0882) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 802.577, TIME@all 0.319 -epoch: [198/350][20/50] time 0.328 (0.331) data 0.001 (0.013) eta 0:42:08 loss 1.1038 (1.0802) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:40:44 loss 1.0817 (1.0878) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 802.496, TIME@all 0.319 -epoch: [198/350][20/50] time 0.328 (0.331) data 0.000 (0.013) eta 0:42:08 loss 1.1007 (1.0734) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:40:45 loss 1.0859 (1.0841) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 802.371, TIME@all 0.319 -epoch: [198/350][20/50] time 0.329 (0.331) data 0.000 (0.013) eta 0:42:09 loss 1.0897 (1.0741) acc 100.0000 (100.0000) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:45 loss 1.0615 (1.0808) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 802.359, TIME@all 0.319 -epoch: [198/350][20/50] time 0.328 (0.331) data 0.001 (0.013) eta 0:42:06 loss 1.1104 (1.0751) acc 100.0000 (99.8438) lr 0.026000 -epoch: [198/350][40/50] time 0.316 (0.321) data 0.000 (0.007) eta 0:40:44 loss 1.0714 (1.0821) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 802.599, TIME@all 0.319 -epoch: [199/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 0:39:27 loss 1.1758 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.308 (0.313) data 0.001 (0.007) eta 0:39:25 loss 1.1781 (1.0932) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 818.955, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:39:28 loss 1.2542 (1.0821) acc 96.8750 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:39:26 loss 1.2125 (1.0929) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 818.549, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:28 loss 1.2710 (1.0754) acc 96.8750 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.1396 (1.0970) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 818.503, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:29 loss 1.1252 (1.0711) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.0917 (1.0832) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 818.466, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:28 loss 1.1661 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:39:26 loss 1.0994 (1.0861) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.552, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:39:29 loss 1.1515 (1.0774) acc 100.0000 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:39:25 loss 1.1166 (1.0886) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 818.766, TIME@all 0.313 -epoch: [199/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:39:28 loss 1.1334 (1.0713) acc 100.0000 (100.0000) lr 0.026000 -epoch: [199/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:39:25 loss 1.1403 (1.0881) acc 96.8750 (99.6094) lr 0.026000 -FPS@all 818.787, TIME@all 0.313 -epoch: [199/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:39:29 loss 1.2339 (1.0790) acc 96.8750 (99.8438) lr 0.026000 -epoch: [199/350][40/50] time 0.306 (0.313) data 0.000 (0.007) eta 0:39:26 loss 1.1948 (1.0869) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.517, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.013) eta 0:39:26 loss 1.1746 (1.0818) acc 100.0000 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:39:12 loss 1.1316 (1.0973) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.913, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0809 (1.0829) acc 100.0000 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:39:12 loss 1.1079 (1.0947) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 818.862, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0994 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:39:13 loss 1.0788 (1.0846) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.746, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0823 (1.0720) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:39:13 loss 1.1439 (1.0900) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.782, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0816 (1.0902) acc 100.0000 (99.5312) lr 0.026000 -epoch: [200/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:39:12 loss 1.1273 (1.0977) acc 96.8750 (99.5312) lr 0.026000 -FPS@all 818.788, TIME@all 0.313 -epoch: [200/350][20/50] time 0.319 (0.314) data 0.001 (0.013) eta 0:39:26 loss 1.1205 (1.0863) acc 100.0000 (99.8438) lr 0.026000 -epoch: [200/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:39:12 loss 1.1079 (1.0879) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.811, TIME@all 0.313 -epoch: [200/350][20/50] time 0.319 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.1487 (1.0749) acc 100.0000 (100.0000) lr 0.026000 -epoch: [200/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:39:12 loss 1.0620 (1.0884) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.819, TIME@all 0.313 -epoch: [200/350][20/50] time 0.318 (0.314) data 0.000 (0.012) eta 0:39:26 loss 1.0940 (1.0810) acc 100.0000 (99.6875) lr 0.026000 -epoch: [200/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:39:12 loss 1.1006 (1.0904) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.782, TIME@all 0.313 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:38:58 loss 1.0926 (1.0804) acc 100.0000 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0820 (1.0916) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.634, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1669 (1.0810) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0640 (1.0887) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.462, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1348 (1.0839) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0718 (1.0940) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.560, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1551 (1.0756) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.001 (0.007) eta 0:38:53 loss 1.0869 (1.0886) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.504, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:38:59 loss 1.1544 (1.0780) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.001 (0.006) eta 0:38:53 loss 1.0662 (1.0893) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.460, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.0825 (1.0791) acc 100.0000 (99.6875) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0702 (1.0906) acc 100.0000 (99.4531) lr 0.026000 -FPS@all 819.462, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.0969 (1.0880) acc 100.0000 (99.8438) lr 0.026000 -epoch: [201/350][40/50] time 0.307 (0.313) data 0.001 (0.007) eta 0:38:53 loss 1.0702 (1.0922) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.504, TIME@all 0.312 -epoch: [201/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:38:59 loss 1.1848 (1.0840) acc 100.0000 (100.0000) lr 0.026000 -epoch: [201/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:38:53 loss 1.0799 (1.0953) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.523, TIME@all 0.312 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0996 (1.0764) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0822 (1.0810) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.592, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0738 (1.0700) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0837 (1.0766) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.441, TIME@all 0.313 -epoch: [202/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.0800 (1.0752) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0844 (1.0847) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.439, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0706 (1.0758) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0663 (1.0806) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.482, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.0885 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0638 (1.0834) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.500, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:38:41 loss 1.0674 (1.0783) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0771 (1.0834) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.539, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.313) data 0.001 (0.012) eta 0:38:41 loss 1.0720 (1.0761) acc 100.0000 (100.0000) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0655 (1.0852) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.492, TIME@all 0.313 -epoch: [202/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:38:41 loss 1.1347 (1.0747) acc 100.0000 (99.8438) lr 0.026000 -epoch: [202/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:38:40 loss 1.0762 (1.0845) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 818.494, TIME@all 0.313 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1681 (1.0810) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:38:23 loss 1.1034 (1.0879) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.375, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.0832 (1.0664) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0880 (1.0784) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.301, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:31 loss 1.1336 (1.0852) acc 100.0000 (99.6875) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.1688 (1.0903) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.217, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:32 loss 1.1494 (1.0787) acc 96.8750 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.3765 (1.0931) acc 93.7500 (99.6875) lr 0.026000 -FPS@all 819.167, TIME@all 0.313 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1010 (1.0710) acc 100.0000 (100.0000) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.001 (0.006) eta 0:38:24 loss 1.0948 (1.0850) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.257, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:38:31 loss 1.1199 (1.0796) acc 100.0000 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:38:24 loss 1.1465 (1.0937) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.208, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:32 loss 1.0993 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -epoch: [203/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0945 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.228, TIME@all 0.312 -epoch: [203/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:38:31 loss 1.1322 (1.0815) acc 96.8750 (99.6875) lr 0.026000 -epoch: [203/350][40/50] time 0.316 (0.313) data 0.000 (0.006) eta 0:38:24 loss 1.0862 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.290, TIME@all 0.312 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0596 (1.0701) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0725 (1.0823) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.792, TIME@all 0.313 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0689 (1.0760) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0579 (1.0893) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.863, TIME@all 0.313 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:38:11 loss 1.0718 (1.0762) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0669 (1.0871) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.722, TIME@all 0.313 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0568 (1.0692) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0621 (1.0850) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.759, TIME@all 0.313 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0643 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -epoch: [204/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0878 (1.0891) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.690, TIME@all 0.313 -epoch: [204/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:38:12 loss 1.0752 (1.0710) acc 100.0000 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.001 (0.006) eta 0:38:08 loss 1.0666 (1.0848) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 818.754, TIME@all 0.313 -epoch: [204/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:38:11 loss 1.0637 (1.0790) acc 100.0000 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:38:09 loss 1.0547 (1.0843) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.755, TIME@all 0.313 -epoch: [204/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 0:38:12 loss 1.0769 (1.0744) acc 100.0000 (99.8438) lr 0.026000 -epoch: [204/350][40/50] time 0.317 (0.313) data 0.000 (0.006) eta 0:38:08 loss 1.0548 (1.0898) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.735, TIME@all 0.313 -epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0637 (1.0697) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.1107 (1.0815) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 817.721, TIME@all 0.313 -epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0908 (1.0674) acc 100.0000 (99.8438) lr 0.026000 -epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:38:01 loss 1.0590 (1.0827) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 817.736, TIME@all 0.313 -epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0628 (1.0630) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.0866 (1.0762) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 817.797, TIME@all 0.313 -epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0860 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.313 (0.314) data 0.001 (0.006) eta 0:38:01 loss 1.0960 (1.0775) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 817.637, TIME@all 0.313 -epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0740 (1.0694) acc 100.0000 (99.8438) lr 0.026000 -epoch: [205/350][40/50] time 0.314 (0.314) data 0.001 (0.006) eta 0:38:01 loss 1.0916 (1.0778) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 817.692, TIME@all 0.313 -epoch: [205/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:38:05 loss 1.0966 (1.0683) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:38:01 loss 1.0644 (1.0768) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 817.700, TIME@all 0.313 -epoch: [205/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.1020 (1.0743) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:38:01 loss 1.0602 (1.0777) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 817.639, TIME@all 0.313 -epoch: [205/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:38:05 loss 1.0993 (1.0723) acc 100.0000 (100.0000) lr 0.026000 -epoch: [205/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:38:01 loss 1.0654 (1.0801) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 817.644, TIME@all 0.313 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0661 (1.0722) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.0878 (1.0809) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 804.372, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:01 loss 1.0705 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0889 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 804.187, TIME@all 0.318 -epoch: [206/350][20/50] time 0.331 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.0610 (1.0628) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0747 (1.0739) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 804.327, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0743 (1.0762) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:23 loss 1.0822 (1.0831) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 804.264, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:00 loss 1.0706 (1.0731) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.1281 (1.0806) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 804.401, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.000 (0.012) eta 0:39:01 loss 1.0686 (1.0724) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.1103 (1.0847) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 804.311, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.0793 (1.0775) acc 100.0000 (99.8438) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.001 (0.006) eta 0:38:22 loss 1.0635 (1.0861) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 804.475, TIME@all 0.318 -epoch: [206/350][20/50] time 0.332 (0.324) data 0.001 (0.012) eta 0:39:00 loss 1.1042 (1.0681) acc 100.0000 (100.0000) lr 0.026000 -epoch: [206/350][40/50] time 0.323 (0.319) data 0.000 (0.006) eta 0:38:22 loss 1.0919 (1.0802) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 804.349, TIME@all 0.318 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.012) eta 0:37:41 loss 1.1307 (1.0747) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.006) eta 0:37:36 loss 1.0892 (1.0858) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.924, TIME@all 0.314 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.013) eta 0:37:42 loss 1.1756 (1.0825) acc 100.0000 (99.3750) lr 0.026000 -epoch: [207/350][40/50] time 0.323 (0.315) data 0.000 (0.007) eta 0:37:37 loss 1.1178 (1.0969) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 814.627, TIME@all 0.314 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1394 (1.0773) acc 100.0000 (100.0000) lr 0.026000 -epoch: [207/350][40/50] time 0.325 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.0854 (1.0975) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 814.524, TIME@all 0.314 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1168 (1.0732) acc 100.0000 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.0980 (1.0856) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.484, TIME@all 0.314 -epoch: [207/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:37:42 loss 1.1201 (1.0670) acc 96.8750 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.327 (0.315) data 0.000 (0.006) eta 0:37:37 loss 1.1424 (1.0829) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 814.574, TIME@all 0.314 -epoch: [207/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:37:42 loss 1.1784 (1.0762) acc 96.8750 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.326 (0.315) data 0.000 (0.007) eta 0:37:37 loss 1.0983 (1.0831) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.529, TIME@all 0.314 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.012) eta 0:37:40 loss 1.1626 (1.0704) acc 96.8750 (99.8438) lr 0.026000 -epoch: [207/350][40/50] time 0.326 (0.315) data 0.001 (0.006) eta 0:37:36 loss 1.0691 (1.0939) acc 100.0000 (99.3750) lr 0.026000 -FPS@all 814.901, TIME@all 0.314 -epoch: [207/350][20/50] time 0.322 (0.315) data 0.001 (0.013) eta 0:37:40 loss 1.0863 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [207/350][40/50] time 0.326 (0.315) data 0.001 (0.007) eta 0:37:36 loss 1.1196 (1.0818) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 814.877, TIME@all 0.314 -epoch: [208/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0693 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0740 (1.0836) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.153, TIME@all 0.313 -epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0533 (1.0739) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0745 (1.0838) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.232, TIME@all 0.313 -epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:37:11 loss 1.0925 (1.0818) acc 100.0000 (99.6875) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:37:05 loss 1.0903 (1.0913) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 818.127, TIME@all 0.313 -epoch: [208/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0670 (1.0729) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0721 (1.0825) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.056, TIME@all 0.313 -epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0541 (1.0709) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0709 (1.0844) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.135, TIME@all 0.313 -epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.1007 (1.0768) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0749 (1.0866) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 818.118, TIME@all 0.313 -epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0897 (1.0727) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.0657 (1.0844) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 818.086, TIME@all 0.313 -epoch: [208/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:37:11 loss 1.0683 (1.0724) acc 100.0000 (100.0000) lr 0.026000 -epoch: [208/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:37:05 loss 1.1045 (1.0801) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 818.192, TIME@all 0.313 -epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.1159 (1.0750) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0766 (1.0844) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.639, TIME@all 0.314 -epoch: [209/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.0991 (1.0719) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.313 (0.315) data 0.000 (0.006) eta 0:37:01 loss 1.0729 (1.0899) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 814.696, TIME@all 0.314 -epoch: [209/350][20/50] time 0.318 (0.315) data 0.001 (0.012) eta 0:37:11 loss 1.0702 (1.0711) acc 100.0000 (99.6875) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0860 (1.0899) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 814.569, TIME@all 0.314 -epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.011) eta 0:37:12 loss 1.1098 (1.0680) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.0910 (1.0798) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 814.577, TIME@all 0.314 -epoch: [209/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 0:37:11 loss 1.1089 (1.0751) acc 96.8750 (99.5312) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.007) eta 0:37:02 loss 1.0600 (1.0857) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 814.641, TIME@all 0.314 -epoch: [209/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:37:11 loss 1.1456 (1.0739) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.001 (0.006) eta 0:37:02 loss 1.0763 (1.0873) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 814.640, TIME@all 0.314 -epoch: [209/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:37:11 loss 1.0564 (1.0664) acc 100.0000 (99.8438) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.001 (0.007) eta 0:37:02 loss 1.0866 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 814.601, TIME@all 0.314 -epoch: [209/350][20/50] time 0.317 (0.315) data 0.001 (0.012) eta 0:37:11 loss 1.0866 (1.0698) acc 100.0000 (100.0000) lr 0.026000 -epoch: [209/350][40/50] time 0.314 (0.315) data 0.000 (0.006) eta 0:37:02 loss 1.1208 (1.0824) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 814.647, TIME@all 0.314 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0514 (1.0645) acc 100.0000 (99.8438) lr 0.026000 -epoch: [210/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0947 (1.0791) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.172, TIME@all 0.313 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0861 (1.0732) acc 100.0000 (99.8438) lr 0.026000 -epoch: [210/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0997 (1.0850) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.220, TIME@all 0.312 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0728 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0680 (1.0814) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.060, TIME@all 0.313 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0767 (1.0771) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0694 (1.0849) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.101, TIME@all 0.313 -epoch: [210/350][20/50] time 0.318 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0703 (1.0693) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0709 (1.0797) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.080, TIME@all 0.313 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0566 (1.0664) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:34 loss 1.0884 (1.0761) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.122, TIME@all 0.313 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:43 loss 1.0535 (1.0684) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:36:34 loss 1.0724 (1.0800) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.182, TIME@all 0.313 -epoch: [210/350][20/50] time 0.319 (0.314) data 0.000 (0.013) eta 0:36:44 loss 1.0673 (1.0673) acc 100.0000 (100.0000) lr 0.026000 -epoch: [210/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:36:34 loss 1.1013 (1.0808) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.165, TIME@all 0.313 -epoch: [211/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:36:21 loss 1.0937 (1.0717) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0589 (1.0840) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.196, TIME@all 0.312 -epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.0611 (1.0712) acc 100.0000 (99.8438) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0634 (1.0801) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.097, TIME@all 0.312 -epoch: [211/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.1027 (1.0717) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0530 (1.0760) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.023, TIME@all 0.312 -epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:22 loss 1.0793 (1.0709) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.0740 (1.0847) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.003, TIME@all 0.312 -epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:36:21 loss 1.0805 (1.0717) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:36:15 loss 1.1023 (1.0845) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.070, TIME@all 0.312 -epoch: [211/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:36:22 loss 1.0646 (1.0718) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0578 (1.0828) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.039, TIME@all 0.312 -epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:36:21 loss 1.0811 (1.0693) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0754 (1.0831) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.115, TIME@all 0.312 -epoch: [211/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:36:21 loss 1.0698 (1.0740) acc 100.0000 (100.0000) lr 0.026000 -epoch: [211/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:36:15 loss 1.0585 (1.0841) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.089, TIME@all 0.312 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0732 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1264 (1.0895) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.163, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:13 loss 1.1018 (1.0677) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.0856 (1.0868) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 816.112, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:13 loss 1.0760 (1.0743) acc 100.0000 (99.8438) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1284 (1.1000) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 815.989, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:13 loss 1.0769 (1.0730) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1292 (1.0893) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 816.027, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0523 (1.0649) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.1274 (1.0824) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.097, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:12 loss 1.0563 (1.0686) acc 100.0000 (99.8438) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1493 (1.0835) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.115, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:36:12 loss 1.0625 (1.0695) acc 100.0000 (100.0000) lr 0.026000 -epoch: [212/350][40/50] time 0.322 (0.314) data 0.000 (0.006) eta 0:36:10 loss 1.1716 (1.0962) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.031, TIME@all 0.314 -epoch: [212/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:36:12 loss 1.0726 (1.0749) acc 100.0000 (99.8438) lr 0.026000 -epoch: [212/350][40/50] time 0.323 (0.314) data 0.000 (0.007) eta 0:36:10 loss 1.2003 (1.0973) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 816.035, TIME@all 0.314 -epoch: [213/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:35:56 loss 1.1226 (1.0787) acc 100.0000 (99.8438) lr 0.026000 -epoch: [213/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.1145 (1.0847) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.148, TIME@all 0.312 -epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0834 (1.0704) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:35:45 loss 1.1028 (1.0861) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.166, TIME@all 0.312 -epoch: [213/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0854 (1.0709) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0706 (1.0782) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.039, TIME@all 0.312 -epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0746 (1.0666) acc 100.0000 (100.0000) lr 0.026000 -epoch: [213/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0644 (1.0812) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.047, TIME@all 0.312 -epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0529 (1.0794) acc 100.0000 (99.8438) lr 0.026000 -epoch: [213/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:35:45 loss 1.0677 (1.0861) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.088, TIME@all 0.312 -epoch: [213/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:35:54 loss 1.0753 (1.0880) acc 100.0000 (99.6875) lr 0.026000 -epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.1020 (1.0926) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 820.055, TIME@all 0.312 -epoch: [213/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:35:54 loss 1.1170 (1.0875) acc 100.0000 (99.6875) lr 0.026000 -epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:35:45 loss 1.0815 (1.0875) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 820.102, TIME@all 0.312 -epoch: [213/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:35:54 loss 1.0999 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -epoch: [213/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:35:45 loss 1.0686 (1.0837) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 820.066, TIME@all 0.312 -epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1259 (1.0778) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1293 (1.0896) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 820.046, TIME@all 0.312 -epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:35:31 loss 1.0950 (1.0706) acc 100.0000 (99.8438) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1587 (1.0906) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.911, TIME@all 0.312 -epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1052 (1.0870) acc 100.0000 (99.6875) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:30 loss 1.0808 (1.0970) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.895, TIME@all 0.312 -epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.0807 (1.0739) acc 100.0000 (99.8438) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1208 (1.0931) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.924, TIME@all 0.312 -epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.2055 (1.0825) acc 96.8750 (99.6875) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1016 (1.0968) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.953, TIME@all 0.312 -epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1763 (1.0815) acc 100.0000 (100.0000) lr 0.026000 -epoch: [214/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1137 (1.1012) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.896, TIME@all 0.312 -epoch: [214/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.0656 (1.0763) acc 100.0000 (99.8438) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.1402 (1.0981) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.956, TIME@all 0.312 -epoch: [214/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:35:31 loss 1.1417 (1.0795) acc 100.0000 (99.6875) lr 0.026000 -epoch: [214/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:29 loss 1.0746 (1.0882) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.978, TIME@all 0.312 -epoch: [215/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:35:25 loss 1.0658 (1.0776) acc 100.0000 (99.8438) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.007) eta 0:35:23 loss 1.1430 (1.0855) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.557, TIME@all 0.314 -epoch: [215/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:35:26 loss 1.1074 (1.0759) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0877 (1.0863) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.413, TIME@all 0.314 -epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1706 (1.0853) acc 96.8750 (99.6875) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0865 (1.0865) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.357, TIME@all 0.314 -epoch: [215/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1601 (1.0801) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0905 (1.0851) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 816.412, TIME@all 0.314 -epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.0970 (1.0758) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:24 loss 1.0928 (1.0905) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 816.467, TIME@all 0.314 -epoch: [215/350][20/50] time 0.317 (0.314) data 0.001 (0.013) eta 0:35:25 loss 1.0839 (1.0745) acc 100.0000 (99.6875) lr 0.026000 -epoch: [215/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:35:24 loss 1.0830 (1.0861) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.422, TIME@all 0.314 -epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.1414 (1.0796) acc 100.0000 (100.0000) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:23 loss 1.0730 (1.0879) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.480, TIME@all 0.314 -epoch: [215/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:35:25 loss 1.0676 (1.0741) acc 100.0000 (99.6875) lr 0.026000 -epoch: [215/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:35:23 loss 1.0796 (1.0855) acc 100.0000 (99.5312) lr 0.026000 -FPS@all 816.470, TIME@all 0.314 -epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:35:10 loss 1.0997 (1.0650) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:59 loss 1.0779 (1.0899) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.378, TIME@all 0.312 -epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1235 (1.0687) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.1766 (1.0841) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.252, TIME@all 0.312 -epoch: [216/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1103 (1.0746) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.1160 (1.0850) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.274, TIME@all 0.312 -epoch: [216/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:35:11 loss 1.1132 (1.0701) acc 100.0000 (99.8438) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:59 loss 1.1453 (1.0894) acc 96.8750 (99.4531) lr 0.026000 -FPS@all 819.342, TIME@all 0.312 -epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:35:11 loss 1.0650 (1.0742) acc 100.0000 (99.6875) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:34:59 loss 1.0754 (1.0852) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.327, TIME@all 0.312 -epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1218 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:34:59 loss 1.1002 (1.0850) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.309, TIME@all 0.312 -epoch: [216/350][20/50] time 0.311 (0.314) data 0.001 (0.012) eta 0:35:11 loss 1.1216 (1.0748) acc 100.0000 (99.8438) lr 0.026000 -epoch: [216/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:35:00 loss 1.0931 (1.0824) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.317, TIME@all 0.312 -epoch: [216/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:35:11 loss 1.1444 (1.0760) acc 100.0000 (100.0000) lr 0.026000 -epoch: [216/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:34:59 loss 1.0942 (1.0943) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.336, TIME@all 0.312 -epoch: [217/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:34:47 loss 1.1087 (1.0742) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:41 loss 1.1219 (1.0859) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.505, TIME@all 0.312 -epoch: [217/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.1404 (1.0808) acc 100.0000 (99.6875) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.0999 (1.0884) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.340, TIME@all 0.312 -epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1746 (1.0717) acc 96.8750 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:42 loss 1.1367 (1.0832) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.317, TIME@all 0.312 -epoch: [217/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:34:47 loss 1.0993 (1.0750) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.1665 (1.0836) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.418, TIME@all 0.312 -epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1177 (1.0782) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:34:42 loss 1.2443 (1.0918) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.410, TIME@all 0.312 -epoch: [217/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.1002 (1.0733) acc 100.0000 (100.0000) lr 0.026000 -epoch: [217/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.1234 (1.0852) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.388, TIME@all 0.312 -epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:34:47 loss 1.1041 (1.0736) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:42 loss 1.2797 (1.0926) acc 93.7500 (99.6094) lr 0.026000 -FPS@all 819.357, TIME@all 0.312 -epoch: [217/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:47 loss 1.0811 (1.0789) acc 100.0000 (99.8438) lr 0.026000 -epoch: [217/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:42 loss 1.0862 (1.0931) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.326, TIME@all 0.312 -epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0983 (1.0684) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0568 (1.0825) acc 100.0000 (99.6875) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.199, TIME@all 0.312 -epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.013) eta 0:34:33 loss 1.0764 (1.0724) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.1151 (1.0873) acc 100.0000 (99.8438) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.303, TIME@all 0.312 -epoch: [218/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:34:34 loss 1.0659 (1.0747) acc 100.0000 (99.8438) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:34:29 loss 1.0795 (1.0859) acc 100.0000 (99.6875) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.126, TIME@all 0.312 -epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.1033 (1.0710) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:29 loss 1.0869 (1.0772) acc 100.0000 (99.8438) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.096, TIME@all 0.312 -epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0567 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:29 loss 1.0736 (1.0862) acc 100.0000 (100.0000) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.180, TIME@all 0.312 -epoch: [218/350][20/50] time 0.317 (0.313) data 0.001 (0.013) eta 0:34:33 loss 1.0703 (1.0750) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0803 (1.0846) acc 100.0000 (99.8438) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.244, TIME@all 0.312 -epoch: [218/350][20/50] time 0.317 (0.313) data 0.000 (0.013) eta 0:34:33 loss 1.0747 (1.0689) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.0561 (1.0762) acc 100.0000 (99.9219) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.174, TIME@all 0.312 -epoch: [218/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:34:34 loss 1.0535 (1.0670) acc 100.0000 (100.0000) lr 0.026000 -epoch: [218/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:34:28 loss 1.1097 (1.0888) acc 100.0000 (99.7656) lr 0.026000 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 16384.0 -FPS@all 820.200, TIME@all 0.312 -epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:18 loss 1.0585 (1.0712) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:12 loss 1.0922 (1.0899) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.685, TIME@all 0.312 -epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0615 (1.0667) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:34:13 loss 1.0571 (1.0777) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.530, TIME@all 0.312 -epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:18 loss 1.0594 (1.0661) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0776 (1.0834) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.621, TIME@all 0.312 -epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0775 (1.0691) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0570 (1.0848) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 819.492, TIME@all 0.312 -epoch: [219/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0661 (1.0775) acc 100.0000 (99.6875) lr 0.026000 -epoch: [219/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:34:13 loss 1.0620 (1.0886) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.516, TIME@all 0.312 -epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0550 (1.0696) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:13 loss 1.0698 (1.0827) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.606, TIME@all 0.312 -epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:34:19 loss 1.0714 (1.0732) acc 100.0000 (99.8438) lr 0.026000 -epoch: [219/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:34:13 loss 1.1110 (1.0904) acc 100.0000 (99.6094) lr 0.026000 -FPS@all 819.557, TIME@all 0.312 -epoch: [219/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:34:19 loss 1.0643 (1.0680) acc 100.0000 (100.0000) lr 0.026000 -epoch: [219/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:13 loss 1.0570 (1.0890) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.546, TIME@all 0.312 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:01 loss 1.0582 (1.0694) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:33:59 loss 1.0712 (1.0795) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 817.511, TIME@all 0.313 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.0837 (1.0672) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0869 (1.0769) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 817.424, TIME@all 0.313 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:34:02 loss 1.0750 (1.0688) acc 100.0000 (99.8438) lr 0.026000 -epoch: [220/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0812 (1.0818) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.336, TIME@all 0.313 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.1204 (1.0736) acc 96.8750 (99.8438) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.0927 (1.0847) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 817.386, TIME@all 0.313 -epoch: [220/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:34:02 loss 1.0862 (1.0752) acc 100.0000 (99.6875) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:00 loss 1.0966 (1.0847) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.344, TIME@all 0.313 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.001 (0.013) eta 0:34:02 loss 1.0904 (1.0707) acc 100.0000 (99.8438) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:34:00 loss 1.1971 (1.0869) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 817.408, TIME@all 0.313 -epoch: [220/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:34:02 loss 1.0768 (1.0736) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:34:00 loss 1.0893 (1.0833) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 817.461, TIME@all 0.313 -epoch: [220/350][20/50] time 0.308 (0.313) data 0.000 (0.013) eta 0:34:02 loss 1.0990 (1.0712) acc 100.0000 (100.0000) lr 0.026000 -epoch: [220/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:34:00 loss 1.0954 (1.0817) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 817.446, TIME@all 0.313 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:33:42 loss 1.1429 (1.0722) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:33:36 loss 1.1264 (1.0795) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.188, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1087 (1.0767) acc 100.0000 (99.8438) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0816 (1.0770) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.098, TIME@all 0.312 -epoch: [221/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1168 (1.0750) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.1172 (1.0839) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 821.025, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:43 loss 1.1077 (1.0813) acc 100.0000 (99.6875) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0750 (1.0809) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 820.972, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.1582 (1.0761) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.1564 (1.0820) acc 96.8750 (99.8438) lr 0.026000 -FPS@all 821.049, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:33:43 loss 1.1509 (1.0875) acc 100.0000 (99.6875) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:33:36 loss 1.0745 (1.0810) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 821.068, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:43 loss 1.1519 (1.0763) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0887 (1.0801) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 821.107, TIME@all 0.312 -epoch: [221/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:33:42 loss 1.0606 (1.0716) acc 100.0000 (100.0000) lr 0.026000 -epoch: [221/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:33:36 loss 1.0957 (1.0771) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 821.092, TIME@all 0.312 -epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0704 (1.0706) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1691 (1.0840) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.776, TIME@all 0.312 -epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0653 (1.0679) acc 100.0000 (99.8438) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0622 (1.0749) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.677, TIME@all 0.312 -epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0562 (1.0676) acc 100.0000 (99.8438) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0748 (1.0762) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.859, TIME@all 0.312 -epoch: [222/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0943 (1.0719) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1034 (1.0826) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.725, TIME@all 0.312 -epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0714 (1.0720) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0586 (1.0834) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.715, TIME@all 0.312 -epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.011) eta 0:33:31 loss 1.0979 (1.0682) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0700 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.676, TIME@all 0.312 -epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:33:30 loss 1.0584 (1.0700) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.1096 (1.0839) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.713, TIME@all 0.312 -epoch: [222/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:33:31 loss 1.0903 (1.0710) acc 100.0000 (100.0000) lr 0.026000 -epoch: [222/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:33:23 loss 1.0625 (1.0768) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 819.714, TIME@all 0.312 -epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:33:22 loss 1.0601 (1.0672) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:33:18 loss 1.0710 (1.0757) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.411, TIME@all 0.314 -epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0949 (1.0685) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1656 (1.0861) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.211, TIME@all 0.314 -epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0915 (1.0675) acc 96.8750 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1161 (1.0791) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 816.318, TIME@all 0.314 -epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0590 (1.0662) acc 100.0000 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.0918 (1.0824) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 816.314, TIME@all 0.314 -epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.1076 (1.0733) acc 100.0000 (99.8438) lr 0.026000 -epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1319 (1.0840) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 816.260, TIME@all 0.314 -epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0790 (1.0666) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1513 (1.0858) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 816.322, TIME@all 0.314 -epoch: [223/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:33:22 loss 1.0642 (1.0639) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:33:19 loss 1.1176 (1.0825) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 816.275, TIME@all 0.314 -epoch: [223/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:33:22 loss 1.0725 (1.0676) acc 100.0000 (100.0000) lr 0.026000 -epoch: [223/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:33:19 loss 1.1335 (1.0806) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 816.314, TIME@all 0.314 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0632 (1.0634) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1048 (1.0781) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.848, TIME@all 0.312 -epoch: [224/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0695 (1.0656) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.0905 (1.0817) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.698, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0706 (1.0688) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1221 (1.0792) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 819.742, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0814 (1.0770) acc 100.0000 (99.6875) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1817 (1.0850) acc 96.8750 (99.6875) lr 0.026000 -FPS@all 819.728, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0972 (1.0725) acc 100.0000 (100.0000) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1449 (1.0843) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.717, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0755 (1.0741) acc 100.0000 (99.8438) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:32:54 loss 1.1502 (1.0899) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 819.683, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.1541 (1.0775) acc 96.8750 (99.6875) lr 0.026000 -epoch: [224/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.1221 (1.0813) acc 96.8750 (99.7656) lr 0.026000 -FPS@all 819.696, TIME@all 0.312 -epoch: [224/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:33:01 loss 1.0697 (1.0689) acc 100.0000 (99.8438) lr 0.026000 -epoch: [224/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:32:54 loss 1.0827 (1.0818) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 819.675, TIME@all 0.312 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.1334 (1.0660) acc 96.8750 (99.8438) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.1423 (1.0792) acc 100.0000 (99.9219) lr 0.026000 -FPS@all 816.043, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0554 (1.0740) acc 100.0000 (99.8438) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.1159 (1.0868) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 815.903, TIME@all 0.314 -epoch: [225/350][20/50] time 0.315 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0554 (1.0681) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0967 (1.0874) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 815.904, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.1417 (1.0702) acc 96.8750 (99.8438) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.0725 (1.0826) acc 100.0000 (99.6875) lr 0.026000 -FPS@all 815.888, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.0617 (1.0651) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0895 (1.0882) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 815.968, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.0692 (1.0631) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:32:46 loss 1.0752 (1.0812) acc 100.0000 (99.7656) lr 0.026000 -FPS@all 815.926, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:32:57 loss 1.1022 (1.0635) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0824 (1.0735) acc 100.0000 (100.0000) lr 0.026000 -FPS@all 815.928, TIME@all 0.314 -epoch: [225/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:32:57 loss 1.0572 (1.0635) acc 100.0000 (100.0000) lr 0.026000 -epoch: [225/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:32:46 loss 1.0584 (1.0789) acc 100.0000 (99.8438) lr 0.026000 -FPS@all 815.959, TIME@all 0.314 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0864 (1.0690) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.1441 (1.0789) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.707, TIME@all 0.313 -epoch: [226/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.1636 (1.0702) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0812 (1.0806) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.663, TIME@all 0.313 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:30 loss 1.1490 (1.0737) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:32:25 loss 1.0615 (1.0763) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.614, TIME@all 0.313 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.2220 (1.0693) acc 96.8750 (99.8438) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:25 loss 1.0558 (1.0787) acc 100.0000 (99.5312) lr 0.002600 -FPS@all 818.604, TIME@all 0.313 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0870 (1.0690) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.1048 (1.0803) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.679, TIME@all 0.313 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0989 (1.0696) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0508 (1.0791) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.723, TIME@all 0.313 -epoch: [226/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.0868 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:25 loss 1.0673 (1.0764) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.668, TIME@all 0.313 -epoch: [226/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:32:29 loss 1.1114 (1.0714) acc 100.0000 (100.0000) lr 0.002600 -epoch: [226/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:32:24 loss 1.0863 (1.0809) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.710, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0913 (1.0681) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0627 (1.0703) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.440, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:32:18 loss 1.0829 (1.0687) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0601 (1.0731) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.503, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.1233 (1.0631) acc 96.8750 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0538 (1.0716) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.376, TIME@all 0.313 -epoch: [227/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:32:18 loss 1.0578 (1.0668) acc 100.0000 (99.6875) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:32:10 loss 1.0524 (1.0750) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 818.403, TIME@all 0.313 -epoch: [227/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0761 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0615 (1.0710) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.392, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0612 (1.0641) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0816 (1.0677) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.397, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0829 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:32:10 loss 1.0592 (1.0726) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.418, TIME@all 0.313 -epoch: [227/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:32:18 loss 1.0540 (1.0716) acc 100.0000 (99.6875) lr 0.002600 -epoch: [227/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:32:10 loss 1.0742 (1.0745) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.407, TIME@all 0.313 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:32:05 loss 1.0690 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0824 (1.0701) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 814.315, TIME@all 0.314 -epoch: [228/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0530 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0827 (1.0751) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 814.128, TIME@all 0.314 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.014) eta 0:32:06 loss 1.0547 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.311 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0772 (1.0683) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 814.203, TIME@all 0.314 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0602 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0706 (1.0677) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.200, TIME@all 0.314 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0805 (1.0679) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.1366 (1.0761) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 814.140, TIME@all 0.314 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0871 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:07 loss 1.0771 (1.0763) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 814.231, TIME@all 0.314 -epoch: [228/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0745 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.001 (0.007) eta 0:32:08 loss 1.0914 (1.0717) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 814.212, TIME@all 0.314 -epoch: [228/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:32:06 loss 1.0519 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [228/350][40/50] time 0.312 (0.316) data 0.000 (0.007) eta 0:32:08 loss 1.0503 (1.0687) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.160, TIME@all 0.314 -epoch: [229/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:31:45 loss 1.0635 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0540 (1.0724) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.769, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:31:44 loss 1.1461 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0710 (1.0702) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.800, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.0853 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:36 loss 1.0551 (1.0746) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.726, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.0848 (1.0595) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:36 loss 1.1294 (1.0672) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.677, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.1768 (1.0679) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0591 (1.0740) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.725, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:31:44 loss 1.1334 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0552 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.726, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:31:44 loss 1.1128 (1.0563) acc 96.8750 (99.8438) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:31:35 loss 1.0613 (1.0661) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.769, TIME@all 0.312 -epoch: [229/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:31:44 loss 1.1244 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -epoch: [229/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:31:35 loss 1.0485 (1.0727) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.718, TIME@all 0.312 -epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.013) eta 0:31:23 loss 1.1052 (1.0669) acc 100.0000 (99.8438) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:31:20 loss 1.0944 (1.0769) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.414, TIME@all 0.312 -epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1507 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0668 (1.0749) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.147, TIME@all 0.313 -epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:31:24 loss 1.1880 (1.0819) acc 96.8750 (99.2188) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0757 (1.0858) acc 100.0000 (99.3750) lr 0.002600 -FPS@all 819.174, TIME@all 0.313 -epoch: [230/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:31:24 loss 1.1013 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.1022 (1.0750) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.249, TIME@all 0.312 -epoch: [230/350][20/50] time 0.316 (0.312) data 0.001 (0.012) eta 0:31:24 loss 1.0915 (1.0728) acc 100.0000 (99.8438) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0980 (1.0800) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.237, TIME@all 0.312 -epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1427 (1.0689) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:31:20 loss 1.1011 (1.0762) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.173, TIME@all 0.313 -epoch: [230/350][20/50] time 0.316 (0.313) data 0.001 (0.012) eta 0:31:24 loss 1.1038 (1.0717) acc 100.0000 (99.6875) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0724 (1.0797) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.188, TIME@all 0.313 -epoch: [230/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:31:24 loss 1.1286 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -epoch: [230/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:31:20 loss 1.0859 (1.0741) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.189, TIME@all 0.313 -epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0566 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0510 (1.0744) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.767, TIME@all 0.313 -epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.014) eta 0:31:23 loss 1.0552 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0903 (1.0735) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.849, TIME@all 0.313 -epoch: [231/350][20/50] time 0.326 (0.315) data 0.000 (0.012) eta 0:31:23 loss 1.0529 (1.0651) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:31:12 loss 1.0590 (1.0737) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.733, TIME@all 0.313 -epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0534 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0935 (1.0768) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.699, TIME@all 0.313 -epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0885 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0492 (1.0717) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.743, TIME@all 0.313 -epoch: [231/350][20/50] time 0.327 (0.315) data 0.001 (0.013) eta 0:31:23 loss 1.0542 (1.0609) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0504 (1.0743) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.724, TIME@all 0.313 -epoch: [231/350][20/50] time 0.327 (0.315) data 0.000 (0.013) eta 0:31:23 loss 1.0637 (1.0650) acc 100.0000 (99.8438) lr 0.002600 -epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:12 loss 1.0489 (1.0710) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.753, TIME@all 0.313 -epoch: [231/350][20/50] time 0.326 (0.315) data 0.001 (0.013) eta 0:31:23 loss 1.0529 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [231/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:31:11 loss 1.0735 (1.0767) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.786, TIME@all 0.313 -epoch: [232/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0776 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0637 (1.0645) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.869, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:30:47 loss 1.0784 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:44 loss 1.0504 (1.0656) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.966, TIME@all 0.312 -epoch: [232/350][20/50] time 0.309 (0.311) data 0.000 (0.013) eta 0:30:47 loss 1.0960 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0599 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.815, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:30:47 loss 1.0588 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:30:45 loss 1.0643 (1.0649) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.800, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0842 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.1056 (1.0657) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 820.851, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0628 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0697 (1.0657) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.804, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:30:47 loss 1.0752 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0793 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.883, TIME@all 0.312 -epoch: [232/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:30:47 loss 1.0643 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [232/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:30:45 loss 1.0807 (1.0615) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.819, TIME@all 0.312 -epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:30:43 loss 1.0645 (1.0657) acc 100.0000 (99.8438) lr 0.002600 -epoch: [233/350][40/50] time 0.320 (0.314) data 0.000 (0.007) eta 0:30:40 loss 1.0629 (1.0768) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.110, TIME@all 0.313 -epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:30:43 loss 1.0599 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.007) eta 0:30:40 loss 1.0502 (1.0735) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.190, TIME@all 0.313 -epoch: [233/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:30:43 loss 1.0813 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0581 (1.0727) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.080, TIME@all 0.313 -epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0632 (1.0696) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0603 (1.0758) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.046, TIME@all 0.313 -epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0710 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0588 (1.0700) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.072, TIME@all 0.313 -epoch: [233/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0565 (1.0611) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0515 (1.0762) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.120, TIME@all 0.313 -epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0899 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0572 (1.0722) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.116, TIME@all 0.313 -epoch: [233/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:30:43 loss 1.0645 (1.0637) acc 100.0000 (99.8438) lr 0.002600 -epoch: [233/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:30:40 loss 1.0772 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.068, TIME@all 0.313 -epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0614 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0527 (1.0633) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.062, TIME@all 0.313 -epoch: [234/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:30:25 loss 1.0554 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0502 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.011, TIME@all 0.313 -epoch: [234/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:30:25 loss 1.0573 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:30:19 loss 1.0664 (1.0668) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.865, TIME@all 0.313 -epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0608 (1.0625) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0593 (1.0714) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.877, TIME@all 0.313 -epoch: [234/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:30:25 loss 1.0682 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0550 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.917, TIME@all 0.313 -epoch: [234/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0657 (1.0617) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0735 (1.0713) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.903, TIME@all 0.313 -epoch: [234/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:30:25 loss 1.0578 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0575 (1.0711) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.954, TIME@all 0.313 -epoch: [234/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:30:25 loss 1.0528 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [234/350][40/50] time 0.309 (0.313) data 0.000 (0.007) eta 0:30:19 loss 1.0512 (1.0665) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.919, TIME@all 0.313 -epoch: [235/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.1035 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0590 (1.0637) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.936, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.0993 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.0794 (1.0698) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.973, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:09 loss 1.0429 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0715 (1.0668) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.815, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:09 loss 1.0643 (1.0546) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0560 (1.0614) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.784, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:08 loss 1.0890 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.1042 (1.0640) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.901, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:08 loss 1.0774 (1.0571) acc 100.0000 (99.8438) lr 0.002600 -epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0714 (1.0697) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.870, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:30:09 loss 1.1745 (1.0635) acc 100.0000 (99.8438) lr 0.002600 -epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:29:59 loss 1.1166 (1.0726) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.858, TIME@all 0.312 -epoch: [235/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:30:08 loss 1.0870 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [235/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:29:59 loss 1.0691 (1.0668) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.866, TIME@all 0.312 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:57 loss 1.1160 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:29:51 loss 1.0574 (1.0717) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.991, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0989 (1.0606) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0562 (1.0762) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.902, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0621 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:29:52 loss 1.0661 (1.0697) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.839, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:29:58 loss 1.0691 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.313 (0.314) data 0.000 (0.006) eta 0:29:52 loss 1.0598 (1.0701) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.800, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0860 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0527 (1.0712) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.846, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0816 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.1109 (1.0727) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.836, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0620 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:29:51 loss 1.0488 (1.0738) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.845, TIME@all 0.313 -epoch: [236/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:29:58 loss 1.0695 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -epoch: [236/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:29:51 loss 1.0720 (1.0747) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.912, TIME@all 0.313 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.015) eta 0:29:35 loss 1.1694 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.314 (0.312) data 0.001 (0.008) eta 0:29:27 loss 1.1569 (1.0776) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.399, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.1039 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0566 (1.0696) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.366, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:29:35 loss 1.0868 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.1290 (1.0726) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 821.200, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:29:35 loss 1.1174 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0885 (1.0746) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.281, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:29:35 loss 1.1267 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0752 (1.0757) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 821.342, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.1203 (1.0690) acc 96.8750 (99.6875) lr 0.002600 -epoch: [237/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:29:27 loss 1.0887 (1.0736) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.345, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.0702 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.1399 (1.0767) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.257, TIME@all 0.312 -epoch: [237/350][20/50] time 0.313 (0.313) data 0.000 (0.014) eta 0:29:35 loss 1.0950 (1.0694) acc 100.0000 (100.0000) lr 0.002600 -epoch: [237/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:29:27 loss 1.0656 (1.0780) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.283, TIME@all 0.312 -epoch: [238/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:29:17 loss 1.0765 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:29:13 loss 1.0519 (1.0670) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.992, TIME@all 0.312 -epoch: [238/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0543 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0553 (1.0717) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.788, TIME@all 0.312 -epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0558 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0499 (1.0689) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.828, TIME@all 0.312 -epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0754 (1.0611) acc 100.0000 (99.8438) lr 0.002600 -epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0608 (1.0683) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.917, TIME@all 0.312 -epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0869 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0555 (1.0763) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.857, TIME@all 0.312 -epoch: [238/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:29:18 loss 1.0487 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0513 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.816, TIME@all 0.312 -epoch: [238/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0510 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0613 (1.0707) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.842, TIME@all 0.312 -epoch: [238/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:29:17 loss 1.0607 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [238/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:29:13 loss 1.0833 (1.0693) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.906, TIME@all 0.312 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0579 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0471 (1.0706) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.694, TIME@all 0.313 -epoch: [239/350][20/50] time 0.315 (0.315) data 0.000 (0.013) eta 0:29:16 loss 1.0539 (1.0674) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:29:04 loss 1.0590 (1.0795) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.769, TIME@all 0.313 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0653 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0580 (1.0720) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.578, TIME@all 0.313 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0539 (1.0619) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0658 (1.0708) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.518, TIME@all 0.313 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.013) eta 0:29:17 loss 1.0530 (1.0590) acc 100.0000 (99.8438) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0496 (1.0683) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.616, TIME@all 0.313 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0620 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0505 (1.0704) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.590, TIME@all 0.313 -epoch: [239/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0550 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0551 (1.0714) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.616, TIME@all 0.313 -epoch: [239/350][20/50] time 0.316 (0.315) data 0.000 (0.012) eta 0:29:17 loss 1.0529 (1.0615) acc 100.0000 (100.0000) lr 0.002600 -epoch: [239/350][40/50] time 0.310 (0.314) data 0.000 (0.006) eta 0:29:04 loss 1.0547 (1.0760) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.614, TIME@all 0.313 -epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.1026 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.0943 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.265, TIME@all 0.312 -epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.1136 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:28:44 loss 1.0621 (1.0685) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.291, TIME@all 0.312 -epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0652 (1.0641) acc 100.0000 (99.8438) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.2136 (1.0712) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 819.181, TIME@all 0.313 -epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:28:53 loss 1.0811 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:28:45 loss 1.0928 (1.0699) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.178, TIME@all 0.313 -epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0519 (1.0640) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.1313 (1.0695) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.195, TIME@all 0.313 -epoch: [240/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:28:53 loss 1.0605 (1.0611) acc 100.0000 (100.0000) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:44 loss 1.0693 (1.0686) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.261, TIME@all 0.312 -epoch: [240/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:28:53 loss 1.0798 (1.0690) acc 100.0000 (99.6875) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.0730 (1.0694) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.152, TIME@all 0.313 -epoch: [240/350][20/50] time 0.315 (0.314) data 0.000 (0.013) eta 0:28:53 loss 1.0657 (1.0689) acc 100.0000 (99.6875) lr 0.002600 -epoch: [240/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:28:45 loss 1.1144 (1.0706) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.211, TIME@all 0.312 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0560 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0532 (1.0677) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.518, TIME@all 0.313 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0636 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0563 (1.0633) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.547, TIME@all 0.313 -epoch: [241/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.0608 (1.0625) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.2085 (1.0735) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 817.423, TIME@all 0.313 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.1087 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.0537 (1.0749) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.401, TIME@all 0.313 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:33 loss 1.0602 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0714 (1.0640) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.421, TIME@all 0.313 -epoch: [241/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:28:32 loss 1.0777 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:28:30 loss 1.0624 (1.0695) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.494, TIME@all 0.313 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:32 loss 1.0824 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0744 (1.0726) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.427, TIME@all 0.313 -epoch: [241/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:28:33 loss 1.0559 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [241/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:28:30 loss 1.0750 (1.0680) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.474, TIME@all 0.313 -epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0699 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.314 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.0833 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 813.976, TIME@all 0.315 -epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.012) eta 0:28:35 loss 1.0623 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:27 loss 1.0706 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 814.006, TIME@all 0.314 -epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0772 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.0947 (1.0673) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 813.983, TIME@all 0.315 -epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0771 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:28 loss 1.1099 (1.0679) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 813.888, TIME@all 0.315 -epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.012) eta 0:28:35 loss 1.0609 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:28 loss 1.1502 (1.0672) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 813.853, TIME@all 0.315 -epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0670 (1.0655) acc 100.0000 (99.8438) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:28 loss 1.0785 (1.0722) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 813.889, TIME@all 0.315 -epoch: [242/350][20/50] time 0.309 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0673 (1.0642) acc 100.0000 (99.6875) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.007) eta 0:28:27 loss 1.1062 (1.0642) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 813.904, TIME@all 0.315 -epoch: [242/350][20/50] time 0.310 (0.316) data 0.000 (0.013) eta 0:28:35 loss 1.0585 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [242/350][40/50] time 0.315 (0.316) data 0.000 (0.006) eta 0:28:28 loss 1.0882 (1.0644) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 813.883, TIME@all 0.315 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0626 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0701 (1.0686) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 807.035, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0777 (1.0584) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0675 (1.0661) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 807.078, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.0645 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0590 (1.0707) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 806.922, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.001 (0.012) eta 0:28:42 loss 1.0594 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:27 loss 1.0640 (1.0704) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 807.174, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.0854 (1.0617) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.320 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0543 (1.0684) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 807.028, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0540 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:28 loss 1.0601 (1.0702) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 806.952, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.012) eta 0:28:42 loss 1.1466 (1.0625) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.006) eta 0:28:28 loss 1.0606 (1.0721) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 806.961, TIME@all 0.317 -epoch: [243/350][20/50] time 0.313 (0.320) data 0.000 (0.013) eta 0:28:42 loss 1.0702 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -epoch: [243/350][40/50] time 0.319 (0.319) data 0.000 (0.007) eta 0:28:27 loss 1.0472 (1.0702) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 807.137, TIME@all 0.317 -epoch: [244/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:27:48 loss 1.0561 (1.0573) acc 100.0000 (99.8438) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0536 (1.0639) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.691, TIME@all 0.314 -epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0669 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0855 (1.0623) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.715, TIME@all 0.314 -epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0922 (1.0624) acc 100.0000 (99.8438) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0625 (1.0727) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 815.563, TIME@all 0.314 -epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0681 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0970 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.612, TIME@all 0.314 -epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.011) eta 0:27:48 loss 1.0542 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0568 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.550, TIME@all 0.314 -epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0853 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.319 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0741 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.705, TIME@all 0.314 -epoch: [244/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0495 (1.0530) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.319 (0.314) data 0.001 (0.006) eta 0:27:49 loss 1.0841 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.632, TIME@all 0.314 -epoch: [244/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:48 loss 1.0485 (1.0506) acc 100.0000 (100.0000) lr 0.002600 -epoch: [244/350][40/50] time 0.320 (0.314) data 0.000 (0.006) eta 0:27:49 loss 1.0595 (1.0617) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 815.582, TIME@all 0.314 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:27:31 loss 1.0503 (1.0653) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0629 (1.0725) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.288, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0492 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0972 (1.0719) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.145, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:27:31 loss 1.0704 (1.0642) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0596 (1.0674) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.223, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:31 loss 1.0463 (1.0684) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0643 (1.0694) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.215, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:31 loss 1.0567 (1.0732) acc 100.0000 (99.8438) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.1027 (1.0768) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.148, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:27:31 loss 1.0567 (1.0648) acc 100.0000 (100.0000) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0616 (1.0714) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.160, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0618 (1.0613) acc 100.0000 (99.8438) lr 0.002600 -epoch: [245/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0568 (1.0709) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.106, TIME@all 0.312 -epoch: [245/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:32 loss 1.0577 (1.0676) acc 100.0000 (99.8438) lr 0.002600 -epoch: [245/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:27:24 loss 1.0608 (1.0712) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.135, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0810 (1.0740) acc 100.0000 (99.5312) lr 0.002600 -epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0674 (1.0736) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.936, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:27:15 loss 1.0475 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:27:11 loss 1.0635 (1.0677) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.900, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0716 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.1009 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.884, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0559 (1.0605) acc 100.0000 (99.8438) lr 0.002600 -epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0537 (1.0748) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.949, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0622 (1.0629) acc 100.0000 (99.8438) lr 0.002600 -epoch: [246/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:27:10 loss 1.0773 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.910, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0513 (1.0661) acc 100.0000 (99.8438) lr 0.002600 -epoch: [246/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0723 (1.0700) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.913, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0581 (1.0619) acc 100.0000 (99.8438) lr 0.002600 -epoch: [246/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0792 (1.0687) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.917, TIME@all 0.313 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [246/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:27:15 loss 1.0886 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -epoch: [246/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:27:10 loss 1.0917 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.889, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.0590 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0596 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.556, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:27:08 loss 1.0849 (1.0628) acc 96.8750 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:26:58 loss 1.0649 (1.0737) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.598, TIME@all 0.313 -epoch: [247/350][20/50] time 0.311 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.0834 (1.0593) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0798 (1.0667) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.508, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.011) eta 0:27:09 loss 1.0760 (1.0623) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:59 loss 1.0953 (1.0684) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.467, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.1665 (1.0702) acc 100.0000 (100.0000) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0782 (1.0698) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.533, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:27:09 loss 1.0901 (1.0640) acc 100.0000 (99.8438) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:26:58 loss 1.1353 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.534, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:27:09 loss 1.1310 (1.0664) acc 96.8750 (99.6875) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:58 loss 1.0661 (1.0702) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.543, TIME@all 0.313 -epoch: [247/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:27:09 loss 1.2356 (1.0668) acc 96.8750 (99.6875) lr 0.002600 -epoch: [247/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:26:58 loss 1.0839 (1.0731) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 817.535, TIME@all 0.313 -epoch: [248/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0495 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0495 (1.0708) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 819.149, TIME@all 0.313 -epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0495 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0512 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.027, TIME@all 0.313 -epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0695 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0782 (1.0633) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.073, TIME@all 0.313 -epoch: [248/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0532 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0547 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.985, TIME@all 0.313 -epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0482 (1.0589) acc 100.0000 (99.8438) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0665 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.005, TIME@all 0.313 -epoch: [248/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:26:41 loss 1.0566 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0585 (1.0709) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 819.023, TIME@all 0.313 -epoch: [248/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:26:41 loss 1.0491 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0900 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.095, TIME@all 0.313 -epoch: [248/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:26:41 loss 1.0545 (1.0577) acc 100.0000 (100.0000) lr 0.002600 -epoch: [248/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:38 loss 1.0503 (1.0625) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.029, TIME@all 0.313 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:39 loss 1.1258 (1.0619) acc 96.8750 (99.8438) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0811 (1.0710) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 815.525, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:39 loss 1.0573 (1.0642) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0617 (1.0724) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 815.577, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.1247 (1.0653) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.317 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0547 (1.0661) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.401, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.012) eta 0:26:40 loss 1.0992 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:26:30 loss 1.0611 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.445, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.0710 (1.0602) acc 100.0000 (99.8438) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0521 (1.0699) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 815.489, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.0792 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:26:30 loss 1.0887 (1.0708) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 815.443, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.000 (0.013) eta 0:26:40 loss 1.1630 (1.0690) acc 96.8750 (99.8438) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.001 (0.007) eta 0:26:30 loss 1.0570 (1.0761) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.471, TIME@all 0.314 -epoch: [249/350][20/50] time 0.328 (0.315) data 0.001 (0.013) eta 0:26:39 loss 1.0750 (1.0621) acc 100.0000 (99.8438) lr 0.002600 -epoch: [249/350][40/50] time 0.318 (0.314) data 0.001 (0.006) eta 0:26:30 loss 1.0745 (1.0685) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.524, TIME@all 0.314 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.001 (0.013) eta 0:26:21 loss 1.1160 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:26:10 loss 1.0966 (1.0716) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.619, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0671 (1.0640) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0934 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.535, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0676 (1.0556) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:26:10 loss 1.0834 (1.0700) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.442, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0582 (1.0651) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0705 (1.0747) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 818.466, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0783 (1.0671) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0702 (1.0777) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.497, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:26:21 loss 1.0735 (1.0608) acc 100.0000 (99.8438) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0945 (1.0695) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.509, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.000 (0.012) eta 0:26:21 loss 1.0511 (1.0665) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.1106 (1.0732) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.532, TIME@all 0.313 -epoch: [250/350][20/50] time 0.309 (0.314) data 0.001 (0.012) eta 0:26:21 loss 1.0705 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [250/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:26:10 loss 1.0628 (1.0678) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.463, TIME@all 0.313 -epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.0852 (1.0608) acc 100.0000 (99.6875) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.001 (0.007) eta 0:26:33 loss 1.1059 (1.0727) acc 96.8750 (99.6875) lr 0.002600 -FPS@all 801.881, TIME@all 0.319 -epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.0820 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:33 loss 1.1115 (1.0700) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 801.710, TIME@all 0.319 -epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.012) eta 0:26:34 loss 1.1233 (1.0609) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.318 (0.321) data 0.000 (0.006) eta 0:26:34 loss 1.0634 (1.0735) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 801.401, TIME@all 0.319 -epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:33 loss 1.1088 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:33 loss 1.0907 (1.0720) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 801.657, TIME@all 0.319 -epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1037 (1.0545) acc 96.8750 (99.8438) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.0605 (1.0687) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 801.444, TIME@all 0.319 -epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1820 (1.0577) acc 96.8750 (99.8438) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.1256 (1.0694) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 801.514, TIME@all 0.319 -epoch: [251/350][20/50] time 0.321 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.0911 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.317 (0.321) data 0.000 (0.007) eta 0:26:34 loss 1.0937 (1.0702) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 801.565, TIME@all 0.319 -epoch: [251/350][20/50] time 0.322 (0.320) data 0.000 (0.013) eta 0:26:34 loss 1.1190 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [251/350][40/50] time 0.316 (0.321) data 0.001 (0.007) eta 0:26:34 loss 1.0532 (1.0708) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 801.428, TIME@all 0.319 -epoch: [252/350][20/50] time 0.320 (0.313) data 0.001 (0.013) eta 0:25:44 loss 1.1117 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0465 (1.0656) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.355, TIME@all 0.313 -epoch: [252/350][20/50] time 0.322 (0.313) data 0.000 (0.014) eta 0:25:45 loss 1.0802 (1.0615) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0549 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.070, TIME@all 0.313 -epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:45 loss 1.1015 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:25:41 loss 1.0574 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 816.891, TIME@all 0.313 -epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.012) eta 0:25:45 loss 1.0834 (1.0621) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:25:41 loss 1.0647 (1.0654) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.911, TIME@all 0.313 -epoch: [252/350][20/50] time 0.320 (0.313) data 0.001 (0.013) eta 0:25:44 loss 1.0839 (1.0641) acc 100.0000 (99.8438) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:40 loss 1.0564 (1.0657) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.120, TIME@all 0.313 -epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:44 loss 1.0598 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:25:40 loss 1.0464 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.295, TIME@all 0.313 -epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.013) eta 0:25:45 loss 1.1003 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:25:41 loss 1.0509 (1.0652) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.953, TIME@all 0.313 -epoch: [252/350][20/50] time 0.321 (0.313) data 0.000 (0.012) eta 0:25:45 loss 1.0540 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [252/350][40/50] time 0.314 (0.314) data 0.001 (0.006) eta 0:25:41 loss 1.0502 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.028, TIME@all 0.313 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.1424 (1.0624) acc 96.8750 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:25:18 loss 1.0609 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.411, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.0884 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:25:18 loss 1.0782 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.462, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1351 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0509 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.319, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1193 (1.0604) acc 100.0000 (99.8438) lr 0.002600 -epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:25:18 loss 1.0461 (1.0651) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.326, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.1024 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:18 loss 1.0870 (1.0635) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.412, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.1591 (1.0633) acc 96.8750 (99.6875) lr 0.002600 -epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0628 (1.0687) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.301, TIME@all 0.312 -epoch: [253/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:25:25 loss 1.0671 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:25:18 loss 1.0535 (1.0668) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.389, TIME@all 0.312 -epoch: [253/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:25:25 loss 1.0889 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [253/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:25:19 loss 1.0504 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.332, TIME@all 0.312 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.1083 (1.0600) acc 100.0000 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1139 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.651, TIME@all 0.313 -epoch: [254/350][20/50] time 0.317 (0.314) data 0.000 (0.012) eta 0:25:16 loss 1.0733 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1092 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 816.554, TIME@all 0.314 -epoch: [254/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.0589 (1.0585) acc 100.0000 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1330 (1.0720) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 816.490, TIME@all 0.314 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0827 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1054 (1.0697) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 816.488, TIME@all 0.314 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.0684 (1.0654) acc 100.0000 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.1138 (1.0722) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.557, TIME@all 0.314 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.000 (0.012) eta 0:25:15 loss 1.1135 (1.0624) acc 96.8750 (99.8438) lr 0.002600 -epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.1502 (1.0741) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 816.516, TIME@all 0.314 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0672 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.327 (0.314) data 0.000 (0.006) eta 0:25:12 loss 1.0636 (1.0730) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.535, TIME@all 0.314 -epoch: [254/350][20/50] time 0.315 (0.314) data 0.001 (0.012) eta 0:25:15 loss 1.0532 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [254/350][40/50] time 0.326 (0.314) data 0.000 (0.006) eta 0:25:11 loss 1.0540 (1.0669) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 816.540, TIME@all 0.314 -epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:24:57 loss 1.0626 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0632 (1.0659) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.155, TIME@all 0.312 -epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0523 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0553 (1.0667) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.082, TIME@all 0.312 -epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:24:58 loss 1.0600 (1.0601) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:24:48 loss 1.0695 (1.0635) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.031, TIME@all 0.312 -epoch: [255/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.1020 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0670 (1.0668) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.021, TIME@all 0.312 -epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0673 (1.0612) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0956 (1.0650) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.059, TIME@all 0.312 -epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:57 loss 1.0645 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.0833 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.111, TIME@all 0.312 -epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:57 loss 1.1141 (1.0665) acc 100.0000 (99.8438) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.1512 (1.0690) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 820.161, TIME@all 0.312 -epoch: [255/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:24:58 loss 1.0571 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [255/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:24:48 loss 1.1289 (1.0627) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 820.030, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0828 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0571 (1.0674) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.417, TIME@all 0.312 -epoch: [256/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:24:35 loss 1.0620 (1.0514) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:24:31 loss 1.0583 (1.0679) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.478, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:24:36 loss 1.0954 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0637 (1.0670) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.341, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0821 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0586 (1.0629) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.303, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0675 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0582 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.352, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0529 (1.0532) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0596 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.368, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0760 (1.0599) acc 100.0000 (99.8438) lr 0.002600 -epoch: [256/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:24:31 loss 1.0478 (1.0678) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.409, TIME@all 0.312 -epoch: [256/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:24:36 loss 1.0821 (1.0528) acc 100.0000 (100.0000) lr 0.002600 -epoch: [256/350][40/50] time 0.314 (0.312) data 0.001 (0.006) eta 0:24:31 loss 1.0600 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.365, TIME@all 0.312 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:24:21 loss 1.0877 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0563 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 815.516, TIME@all 0.314 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.0779 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0624 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.394, TIME@all 0.314 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:24:21 loss 1.0566 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.006) eta 0:24:23 loss 1.0780 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.413, TIME@all 0.314 -epoch: [257/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:24:21 loss 1.1451 (1.0599) acc 100.0000 (99.8438) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.006) eta 0:24:23 loss 1.0867 (1.0677) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 815.428, TIME@all 0.314 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.1386 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0712 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.395, TIME@all 0.314 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.1191 (1.0611) acc 96.8750 (99.6875) lr 0.002600 -epoch: [257/350][40/50] time 0.321 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.1240 (1.0669) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 815.458, TIME@all 0.314 -epoch: [257/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 0:24:21 loss 1.0545 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0727 (1.0670) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.429, TIME@all 0.314 -epoch: [257/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:24:21 loss 1.0909 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -epoch: [257/350][40/50] time 0.322 (0.314) data 0.000 (0.007) eta 0:24:23 loss 1.0479 (1.0671) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 815.481, TIME@all 0.314 -epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0594 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0747 (1.0665) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.566, TIME@all 0.312 -epoch: [258/350][20/50] time 0.310 (0.315) data 0.000 (0.013) eta 0:24:16 loss 1.1340 (1.0627) acc 96.8750 (99.5312) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:24:04 loss 1.1445 (1.0709) acc 96.8750 (99.6094) lr 0.002600 -FPS@all 819.596, TIME@all 0.312 -epoch: [258/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0743 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0574 (1.0656) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.485, TIME@all 0.312 -epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.1720 (1.0620) acc 96.8750 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0951 (1.0733) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.470, TIME@all 0.312 -epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0771 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0711 (1.0714) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.536, TIME@all 0.312 -epoch: [258/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.1036 (1.0641) acc 100.0000 (99.8438) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0844 (1.0700) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.531, TIME@all 0.312 -epoch: [258/350][20/50] time 0.311 (0.315) data 0.000 (0.012) eta 0:24:16 loss 1.0610 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:24:04 loss 1.0750 (1.0709) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.518, TIME@all 0.312 -epoch: [258/350][20/50] time 0.311 (0.315) data 0.001 (0.012) eta 0:24:16 loss 1.1189 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [258/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:24:04 loss 1.0748 (1.0695) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.600, TIME@all 0.312 -epoch: [259/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1706 (1.0609) acc 96.8750 (99.8438) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0572 (1.0665) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.678, TIME@all 0.312 -epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1602 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0582 (1.0730) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.593, TIME@all 0.312 -epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:23:49 loss 1.0685 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:23:45 loss 1.0508 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.538, TIME@all 0.312 -epoch: [259/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1129 (1.0563) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0507 (1.0683) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.554, TIME@all 0.312 -epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1173 (1.0568) acc 96.8750 (99.8438) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0715 (1.0684) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.586, TIME@all 0.312 -epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.1403 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.001 (0.007) eta 0:23:45 loss 1.0662 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.527, TIME@all 0.312 -epoch: [259/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:23:49 loss 1.0716 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [259/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0469 (1.0662) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.554, TIME@all 0.312 -epoch: [259/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:23:49 loss 1.1356 (1.0608) acc 96.8750 (99.8438) lr 0.002600 -epoch: [259/350][40/50] time 0.320 (0.313) data 0.000 (0.007) eta 0:23:45 loss 1.0481 (1.0704) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.603, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0691 (1.0622) acc 100.0000 (99.8438) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.1186 (1.0757) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.604, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0615 (1.0581) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0517 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.472, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0641 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0614 (1.0674) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.458, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:23:37 loss 1.0904 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0483 (1.0669) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.487, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:23:37 loss 1.1344 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:23:31 loss 1.0495 (1.0750) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.470, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.1105 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0516 (1.0667) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.464, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.0795 (1.0646) acc 100.0000 (99.8438) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:23:31 loss 1.0617 (1.0749) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 819.558, TIME@all 0.312 -epoch: [260/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:23:37 loss 1.1173 (1.0619) acc 100.0000 (99.8438) lr 0.002600 -epoch: [260/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:23:31 loss 1.0524 (1.0725) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.514, TIME@all 0.312 -epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:23:18 loss 1.0692 (1.0615) acc 100.0000 (99.8438) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:23:13 loss 1.0502 (1.0681) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.083, TIME@all 0.312 -epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0695 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0659 (1.0670) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.127, TIME@all 0.312 -epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0671 (1.0672) acc 100.0000 (99.6875) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0801 (1.0720) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.035, TIME@all 0.312 -epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0844 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0762 (1.0691) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.984, TIME@all 0.312 -epoch: [261/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:23:18 loss 1.0695 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:23:13 loss 1.0735 (1.0630) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.033, TIME@all 0.312 -epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0536 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0920 (1.0732) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.046, TIME@all 0.312 -epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0677 (1.0600) acc 100.0000 (99.8438) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0602 (1.0699) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.071, TIME@all 0.312 -epoch: [261/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:23:18 loss 1.0597 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [261/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:23:13 loss 1.0695 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.035, TIME@all 0.312 -epoch: [262/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:23:13 loss 1.0680 (1.0675) acc 100.0000 (99.6875) lr 0.002600 -epoch: [262/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0673 (1.0730) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 816.949, TIME@all 0.313 -epoch: [262/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:23:13 loss 1.0664 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:23:04 loss 1.0717 (1.0629) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.966, TIME@all 0.313 -epoch: [262/350][20/50] time 0.315 (0.314) data 0.001 (0.011) eta 0:23:13 loss 1.0466 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0574 (1.0643) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 816.874, TIME@all 0.313 -epoch: [262/350][20/50] time 0.314 (0.315) data 0.000 (0.012) eta 0:23:13 loss 1.0533 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0794 (1.0703) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.830, TIME@all 0.313 -epoch: [262/350][20/50] time 0.315 (0.314) data 0.001 (0.013) eta 0:23:13 loss 1.0502 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:23:04 loss 1.1793 (1.0656) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.909, TIME@all 0.313 -epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0553 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0652 (1.0648) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 816.891, TIME@all 0.313 -epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0638 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.001 (0.006) eta 0:23:04 loss 1.0613 (1.0606) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 816.928, TIME@all 0.313 -epoch: [262/350][20/50] time 0.314 (0.314) data 0.001 (0.012) eta 0:23:13 loss 1.0905 (1.0614) acc 100.0000 (99.8438) lr 0.002600 -epoch: [262/350][40/50] time 0.316 (0.314) data 0.000 (0.006) eta 0:23:04 loss 1.0801 (1.0688) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 816.984, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.014) eta 0:22:53 loss 1.0478 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.310 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.1144 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.417, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.012) eta 0:22:54 loss 1.0482 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.006) eta 0:22:47 loss 1.0690 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.223, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:22:54 loss 1.0548 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0658 (1.0661) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.215, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:22:53 loss 1.0811 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0845 (1.0675) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.283, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0760 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.1074 (1.0665) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 818.319, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.014) eta 0:22:53 loss 1.0564 (1.0600) acc 100.0000 (99.8438) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0836 (1.0713) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.278, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0689 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0669 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.288, TIME@all 0.313 -epoch: [263/350][20/50] time 0.311 (0.314) data 0.001 (0.013) eta 0:22:53 loss 1.0503 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [263/350][40/50] time 0.309 (0.314) data 0.000 (0.007) eta 0:22:47 loss 1.0861 (1.0685) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.336, TIME@all 0.313 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0551 (1.0545) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0475 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.299, TIME@all 0.312 -epoch: [264/350][20/50] time 0.314 (0.313) data 0.000 (0.011) eta 0:22:35 loss 1.0502 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0619 (1.0648) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.197, TIME@all 0.313 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0464 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0677 (1.0705) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.158, TIME@all 0.313 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0458 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0559 (1.0680) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.223, TIME@all 0.312 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0570 (1.0594) acc 100.0000 (99.8438) lr 0.002600 -epoch: [264/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0549 (1.0710) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.243, TIME@all 0.312 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0564 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:22:29 loss 1.0995 (1.0695) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.242, TIME@all 0.312 -epoch: [264/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0785 (1.0581) acc 100.0000 (99.8438) lr 0.002600 -epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0504 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.199, TIME@all 0.313 -epoch: [264/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:22:35 loss 1.0548 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [264/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:22:29 loss 1.0454 (1.0715) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.224, TIME@all 0.312 -epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0618 (1.0570) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0516 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.503, TIME@all 0.313 -epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0676 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0492 (1.0628) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.426, TIME@all 0.313 -epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0585 (1.0628) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0523 (1.0653) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.356, TIME@all 0.313 -epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0600 (1.0644) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0542 (1.0650) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.378, TIME@all 0.313 -epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0505 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:22:16 loss 1.0529 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 817.313, TIME@all 0.313 -epoch: [265/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0605 (1.0580) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0548 (1.0674) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.421, TIME@all 0.313 -epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:22:22 loss 1.0611 (1.0662) acc 100.0000 (99.8438) lr 0.002600 -epoch: [265/350][40/50] time 0.316 (0.314) data 0.000 (0.007) eta 0:22:16 loss 1.0521 (1.0734) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 817.364, TIME@all 0.313 -epoch: [265/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:22:22 loss 1.0555 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [265/350][40/50] time 0.315 (0.314) data 0.001 (0.006) eta 0:22:16 loss 1.0515 (1.0684) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 817.406, TIME@all 0.313 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:22:12 loss 1.1244 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.1069 (1.0718) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.605, TIME@all 0.314 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0599 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.1140 (1.0666) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 814.574, TIME@all 0.314 -epoch: [266/350][20/50] time 0.318 (0.315) data 0.000 (0.013) eta 0:22:11 loss 1.0751 (1.0570) acc 100.0000 (99.8438) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.0579 (1.0650) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.663, TIME@all 0.314 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.013) eta 0:22:12 loss 1.1002 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -epoch: [266/350][40/50] time 0.317 (0.315) data 0.000 (0.006) eta 0:22:05 loss 1.0522 (1.0684) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.505, TIME@all 0.314 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0965 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.1032 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 814.578, TIME@all 0.314 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.001 (0.013) eta 0:22:12 loss 1.1078 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.007) eta 0:22:04 loss 1.0783 (1.0676) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 814.596, TIME@all 0.314 -epoch: [266/350][20/50] time 0.318 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0937 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.001 (0.006) eta 0:22:04 loss 1.0824 (1.0755) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 814.607, TIME@all 0.314 -epoch: [266/350][20/50] time 0.317 (0.315) data 0.000 (0.012) eta 0:22:12 loss 1.0874 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [266/350][40/50] time 0.318 (0.315) data 0.000 (0.006) eta 0:22:04 loss 1.0765 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 814.570, TIME@all 0.314 -epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0695 (1.0659) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:21:42 loss 1.0613 (1.0653) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.436, TIME@all 0.313 -epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0742 (1.0608) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:42 loss 1.0592 (1.0627) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.466, TIME@all 0.313 -epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.1049 (1.0601) acc 96.8750 (99.6875) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0636 (1.0676) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 818.368, TIME@all 0.313 -epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0772 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0653 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.294, TIME@all 0.313 -epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0518 (1.0575) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0827 (1.0663) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.353, TIME@all 0.313 -epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:21:47 loss 1.0669 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:21:43 loss 1.0578 (1.0668) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.366, TIME@all 0.313 -epoch: [267/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0691 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.0736 (1.0654) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.302, TIME@all 0.313 -epoch: [267/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:21:47 loss 1.0807 (1.0644) acc 100.0000 (99.8438) lr 0.002600 -epoch: [267/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:21:43 loss 1.1050 (1.0686) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.350, TIME@all 0.313 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0613 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0810 (1.0659) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.840, TIME@all 0.312 -epoch: [268/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:21:30 loss 1.0567 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:25 loss 1.0880 (1.0674) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.902, TIME@all 0.312 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0761 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0644 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.785, TIME@all 0.312 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0495 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.1158 (1.0645) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.783, TIME@all 0.312 -epoch: [268/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:21:31 loss 1.0670 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:21:25 loss 1.0569 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.760, TIME@all 0.312 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:21:30 loss 1.1609 (1.0594) acc 96.8750 (99.8438) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:25 loss 1.0619 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.762, TIME@all 0.312 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0718 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0844 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.815, TIME@all 0.312 -epoch: [268/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:21:30 loss 1.0970 (1.0642) acc 100.0000 (100.0000) lr 0.002600 -epoch: [268/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:21:25 loss 1.0788 (1.0717) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.825, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0573 (1.0604) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.1133 (1.0700) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.623, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:21:19 loss 1.0424 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0952 (1.0706) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 819.686, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:21:19 loss 1.0801 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:21:11 loss 1.0995 (1.0741) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.569, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0838 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0917 (1.0722) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.546, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0562 (1.0549) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.1012 (1.0730) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.568, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.014) eta 0:21:19 loss 1.1331 (1.0623) acc 100.0000 (100.0000) lr 0.002600 -epoch: [269/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0828 (1.0733) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.564, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0806 (1.0638) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0709 (1.0724) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.620, TIME@all 0.312 -epoch: [269/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:21:19 loss 1.0717 (1.0585) acc 100.0000 (99.8438) lr 0.002600 -epoch: [269/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:21:11 loss 1.0773 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.617, TIME@all 0.312 -epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:21:04 loss 1.0617 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0585 (1.0638) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.399, TIME@all 0.313 -epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0463 (1.0606) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0931 (1.0691) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.132, TIME@all 0.313 -epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0555 (1.0577) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.315 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0492 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.170, TIME@all 0.313 -epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0477 (1.0542) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0641 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.142, TIME@all 0.313 -epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0490 (1.0581) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0955 (1.0644) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 818.136, TIME@all 0.313 -epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.011) eta 0:21:05 loss 1.0888 (1.0628) acc 100.0000 (99.8438) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0695 (1.0637) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.085, TIME@all 0.313 -epoch: [270/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0670 (1.0517) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0807 (1.0652) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.141, TIME@all 0.313 -epoch: [270/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:21:04 loss 1.0527 (1.0582) acc 100.0000 (100.0000) lr 0.002600 -epoch: [270/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:20:57 loss 1.0576 (1.0674) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.175, TIME@all 0.313 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:20:46 loss 1.1241 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:39 loss 1.0702 (1.0750) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.792, TIME@all 0.312 -epoch: [271/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1162 (1.0653) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0894 (1.0752) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.639, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0583 (1.0664) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.1296 (1.0753) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.756, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1315 (1.0689) acc 100.0000 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.1048 (1.0793) acc 100.0000 (99.6094) lr 0.002600 -FPS@all 819.742, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.1090 (1.0607) acc 96.8750 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0753 (1.0682) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.635, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:20:46 loss 1.1258 (1.0642) acc 96.8750 (99.8438) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:20:39 loss 1.1463 (1.0707) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 819.666, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0915 (1.0655) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:20:39 loss 1.0688 (1.0763) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.657, TIME@all 0.312 -epoch: [271/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:20:46 loss 1.0746 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -epoch: [271/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:20:39 loss 1.0872 (1.0717) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.686, TIME@all 0.312 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.014) eta 0:20:30 loss 1.0575 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:24 loss 1.1060 (1.0669) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.684, TIME@all 0.313 -epoch: [272/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:20:30 loss 1.0574 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0777 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.592, TIME@all 0.313 -epoch: [272/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:20:30 loss 1.0462 (1.0523) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:20:25 loss 1.0629 (1.0640) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.553, TIME@all 0.313 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:31 loss 1.0642 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0622 (1.0693) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.576, TIME@all 0.313 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:20:31 loss 1.0633 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:20:25 loss 1.0603 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.549, TIME@all 0.313 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:20:30 loss 1.0610 (1.0555) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0566 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.593, TIME@all 0.313 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:31 loss 1.0653 (1.0506) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0546 (1.0643) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.551, TIME@all 0.313 -epoch: [272/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:20:30 loss 1.0518 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [272/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:25 loss 1.0766 (1.0717) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.618, TIME@all 0.313 -epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0759 (1.0619) acc 100.0000 (99.8438) lr 0.002600 -epoch: [273/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0737 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.507, TIME@all 0.312 -epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.1392 (1.0651) acc 100.0000 (99.8438) lr 0.002600 -epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0632 (1.0664) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.347, TIME@all 0.312 -epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.012) eta 0:20:16 loss 1.0945 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:20:07 loss 1.0621 (1.0627) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.330, TIME@all 0.312 -epoch: [273/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:20:16 loss 1.0897 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0706 (1.0731) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.523, TIME@all 0.312 -epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0774 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0532 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.392, TIME@all 0.312 -epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0528 (1.0566) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.310 (0.313) data 0.001 (0.007) eta 0:20:07 loss 1.0507 (1.0687) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.373, TIME@all 0.312 -epoch: [273/350][20/50] time 0.311 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0982 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0783 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.355, TIME@all 0.312 -epoch: [273/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:20:16 loss 1.0677 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [273/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:20:07 loss 1.0610 (1.0630) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.406, TIME@all 0.312 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0670 (1.0604) acc 100.0000 (99.8438) lr 0.002600 -epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0788 (1.0725) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.213, TIME@all 0.312 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:20:04 loss 1.0545 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:19:54 loss 1.0659 (1.0619) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.286, TIME@all 0.312 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0503 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0633 (1.0660) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.232, TIME@all 0.312 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0587 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0555 (1.0608) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.146, TIME@all 0.313 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0682 (1.0626) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0534 (1.0709) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.198, TIME@all 0.313 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0677 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0758 (1.0637) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.193, TIME@all 0.313 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.1912 (1.0649) acc 96.8750 (99.8438) lr 0.002600 -epoch: [274/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0607 (1.0702) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.131, TIME@all 0.313 -epoch: [274/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:20:04 loss 1.0484 (1.0572) acc 100.0000 (100.0000) lr 0.002600 -epoch: [274/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:19:54 loss 1.0498 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.197, TIME@all 0.313 -epoch: [275/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:19:41 loss 1.0695 (1.0611) acc 100.0000 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.0522 (1.0692) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.821, TIME@all 0.312 -epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.1124 (1.0592) acc 100.0000 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.1097 (1.0676) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.696, TIME@all 0.312 -epoch: [275/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.0742 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.1008 (1.0768) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.732, TIME@all 0.312 -epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:19:41 loss 1.1003 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.0675 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.694, TIME@all 0.312 -epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.1078 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:36 loss 1.0603 (1.0720) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.707, TIME@all 0.312 -epoch: [275/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:19:41 loss 1.0933 (1.0527) acc 100.0000 (100.0000) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.1168 (1.0638) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.690, TIME@all 0.312 -epoch: [275/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:19:41 loss 1.0581 (1.0622) acc 100.0000 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:19:36 loss 1.1114 (1.0721) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.696, TIME@all 0.312 -epoch: [275/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:19:41 loss 1.1303 (1.0556) acc 96.8750 (99.8438) lr 0.002600 -epoch: [275/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:19:36 loss 1.0798 (1.0676) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.748, TIME@all 0.312 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0923 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:19:22 loss 1.2261 (1.0761) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.666, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:19:28 loss 1.0837 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0640 (1.0703) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.536, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0616 (1.0611) acc 100.0000 (99.8438) lr 0.002600 -epoch: [276/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:19:22 loss 1.0784 (1.0693) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.541, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0591 (1.0577) acc 100.0000 (99.8438) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0673 (1.0730) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.596, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.001 (0.012) eta 0:19:28 loss 1.0544 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0556 (1.0718) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.563, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0478 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0780 (1.0716) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.565, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:19:28 loss 1.0744 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [276/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:19:22 loss 1.0914 (1.0664) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.551, TIME@all 0.313 -epoch: [276/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:19:28 loss 1.0713 (1.0644) acc 100.0000 (99.8438) lr 0.002600 -epoch: [276/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:19:22 loss 1.0776 (1.0699) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.562, TIME@all 0.313 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0594 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0979 (1.0666) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 822.018, TIME@all 0.311 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0481 (1.0608) acc 100.0000 (99.8438) lr 0.002600 -epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.1277 (1.0738) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 821.925, TIME@all 0.311 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0613 (1.0556) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.0785 (1.0699) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 821.883, TIME@all 0.311 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.001 (0.012) eta 0:19:09 loss 1.0631 (1.0647) acc 100.0000 (99.8438) lr 0.002600 -epoch: [277/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.0530 (1.0717) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.909, TIME@all 0.311 -epoch: [277/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0575 (1.0591) acc 100.0000 (99.8438) lr 0.002600 -epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0754 (1.0680) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 821.949, TIME@all 0.311 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0705 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0967 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.891, TIME@all 0.311 -epoch: [277/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:19:09 loss 1.0577 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:19:02 loss 1.1134 (1.0723) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.928, TIME@all 0.311 -epoch: [277/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:19:09 loss 1.0645 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [277/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:19:02 loss 1.0854 (1.0712) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.893, TIME@all 0.311 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.1210 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0530 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.969, TIME@all 0.312 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:18:59 loss 1.0819 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.307 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0509 (1.0647) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.998, TIME@all 0.312 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.0808 (1.0551) acc 100.0000 (99.8438) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0925 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.858, TIME@all 0.312 -epoch: [278/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:18:59 loss 1.1253 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:18:50 loss 1.0595 (1.0692) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.860, TIME@all 0.312 -epoch: [278/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:18:58 loss 1.0933 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.006) eta 0:18:50 loss 1.0620 (1.0676) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.029, TIME@all 0.312 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.0953 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0628 (1.0668) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.915, TIME@all 0.312 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:18:59 loss 1.1094 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0458 (1.0640) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.889, TIME@all 0.312 -epoch: [278/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:18:59 loss 1.0767 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [278/350][40/50] time 0.308 (0.313) data 0.000 (0.007) eta 0:18:50 loss 1.0618 (1.0681) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.946, TIME@all 0.312 -epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0765 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1777 (1.0664) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.105, TIME@all 0.312 -epoch: [279/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0535 (1.0525) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1228 (1.0627) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.024, TIME@all 0.312 -epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:18:38 loss 1.0546 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0893 (1.0621) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.021, TIME@all 0.312 -epoch: [279/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:38 loss 1.0774 (1.0575) acc 100.0000 (99.8438) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0872 (1.0624) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.052, TIME@all 0.312 -epoch: [279/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:18:38 loss 1.0975 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:18:31 loss 1.0819 (1.0667) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.184, TIME@all 0.312 -epoch: [279/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:18:38 loss 1.0859 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0964 (1.0647) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.061, TIME@all 0.312 -epoch: [279/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0646 (1.0531) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.1593 (1.0637) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 821.109, TIME@all 0.312 -epoch: [279/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:18:38 loss 1.0609 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [279/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:31 loss 1.0751 (1.0607) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.087, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0521 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0577 (1.0743) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.749, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:18:21 loss 1.0596 (1.0615) acc 100.0000 (99.8438) lr 0.002600 -epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0569 (1.0703) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.815, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0778 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0565 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.681, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0571 (1.0535) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0599 (1.0657) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.711, TIME@all 0.312 -epoch: [280/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:18:21 loss 1.0560 (1.0581) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:18:15 loss 1.0485 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.682, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.0580 (1.0522) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0540 (1.0675) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.730, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:18:21 loss 1.0631 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0545 (1.0622) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.767, TIME@all 0.312 -epoch: [280/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:18:21 loss 1.1517 (1.0622) acc 96.8750 (99.6875) lr 0.002600 -epoch: [280/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:18:15 loss 1.0537 (1.0698) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.713, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0469 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0628 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.239, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:07 loss 1.0633 (1.0662) acc 100.0000 (100.0000) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0457 (1.0681) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.311, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0559 (1.0644) acc 100.0000 (99.6875) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0610 (1.0667) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.142, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0591 (1.0623) acc 100.0000 (99.6875) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0541 (1.0717) acc 100.0000 (99.3750) lr 0.002600 -FPS@all 820.157, TIME@all 0.312 -epoch: [281/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:07 loss 1.1096 (1.0620) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:18:02 loss 1.0477 (1.0698) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.266, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0951 (1.0616) acc 96.8750 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0447 (1.0700) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.169, TIME@all 0.312 -epoch: [281/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0570 (1.0601) acc 100.0000 (99.8438) lr 0.002600 -epoch: [281/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:18:02 loss 1.0536 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.212, TIME@all 0.312 -epoch: [281/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:18:08 loss 1.0600 (1.0605) acc 100.0000 (100.0000) lr 0.002600 -epoch: [281/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:18:02 loss 1.0842 (1.0700) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.188, TIME@all 0.312 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1194 (1.0590) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.001 (0.007) eta 0:17:42 loss 1.0724 (1.0677) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.941, TIME@all 0.311 -epoch: [282/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.0936 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:41 loss 1.0569 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.983, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:49 loss 1.1055 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0653 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.825, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:49 loss 1.0928 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0479 (1.0635) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 822.828, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.0565 (1.0557) acc 100.0000 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0462 (1.0635) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.889, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:17:48 loss 1.0620 (1.0608) acc 100.0000 (99.8438) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:17:42 loss 1.0525 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.902, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1407 (1.0634) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0713 (1.0668) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 822.890, TIME@all 0.311 -epoch: [282/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:17:48 loss 1.1288 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [282/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:17:42 loss 1.0735 (1.0643) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 822.897, TIME@all 0.311 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0518 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0651 (1.0669) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.229, TIME@all 0.312 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.1001 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0567 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.254, TIME@all 0.312 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0466 (1.0562) acc 100.0000 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0565 (1.0704) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.075, TIME@all 0.312 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0579 (1.0552) acc 100.0000 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0935 (1.0658) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.112, TIME@all 0.312 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.011) eta 0:17:36 loss 1.0707 (1.0560) acc 100.0000 (99.8438) lr 0.002600 -epoch: [283/350][40/50] time 0.309 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.1615 (1.0706) acc 96.8750 (99.6094) lr 0.002600 -FPS@all 820.099, TIME@all 0.312 -epoch: [283/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0523 (1.0547) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0554 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.173, TIME@all 0.312 -epoch: [283/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0861 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:17:30 loss 1.0867 (1.0699) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.208, TIME@all 0.312 -epoch: [283/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:17:36 loss 1.0612 (1.0530) acc 100.0000 (100.0000) lr 0.002600 -epoch: [283/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:17:30 loss 1.0764 (1.0644) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.135, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1092 (1.0571) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:14 loss 1.1271 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.465, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:17:21 loss 1.0694 (1.0610) acc 100.0000 (99.8438) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:17:14 loss 1.1863 (1.0710) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 819.471, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.2068 (1.0728) acc 96.8750 (99.6875) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.2257 (1.0795) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 819.347, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1417 (1.0599) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1486 (1.0671) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.367, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.0849 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1821 (1.0745) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 819.408, TIME@all 0.312 -epoch: [284/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1134 (1.0574) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:17:14 loss 1.1131 (1.0753) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.430, TIME@all 0.312 -epoch: [284/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.0892 (1.0607) acc 100.0000 (99.6875) lr 0.002600 -epoch: [284/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1210 (1.0711) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.349, TIME@all 0.312 -epoch: [284/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:17:21 loss 1.1210 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -epoch: [284/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:17:15 loss 1.1240 (1.0779) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 819.375, TIME@all 0.312 -epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.1203 (1.0641) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:16:57 loss 1.0450 (1.0704) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.557, TIME@all 0.312 -epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:17:05 loss 1.1627 (1.0613) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:58 loss 1.0519 (1.0693) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.446, TIME@all 0.312 -epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0890 (1.0629) acc 100.0000 (99.6875) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:16:58 loss 1.0560 (1.0732) acc 100.0000 (99.5312) lr 0.002600 -FPS@all 820.424, TIME@all 0.312 -epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0964 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:16:57 loss 1.0795 (1.0669) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.498, TIME@all 0.312 -epoch: [285/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0668 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [285/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:16:57 loss 1.0513 (1.0666) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.491, TIME@all 0.312 -epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.1674 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:58 loss 1.0611 (1.0703) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.462, TIME@all 0.312 -epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:17:05 loss 1.0627 (1.0648) acc 100.0000 (99.8438) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:16:58 loss 1.0511 (1.0669) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.448, TIME@all 0.312 -epoch: [285/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:17:05 loss 1.2219 (1.0689) acc 93.7500 (99.6875) lr 0.002600 -epoch: [285/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:16:57 loss 1.0556 (1.0706) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.484, TIME@all 0.312 -epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.014) eta 0:16:48 loss 1.0589 (1.0507) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0555 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.986, TIME@all 0.312 -epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1288 (1.0586) acc 100.0000 (99.6875) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0534 (1.0629) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.822, TIME@all 0.312 -epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:16:48 loss 1.0799 (1.0533) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0492 (1.0644) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.820, TIME@all 0.312 -epoch: [286/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.0965 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0689 (1.0635) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.885, TIME@all 0.312 -epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1324 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0533 (1.0667) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.905, TIME@all 0.312 -epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.0627 (1.0525) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0477 (1.0611) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.870, TIME@all 0.312 -epoch: [286/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:16:48 loss 1.1191 (1.0616) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:16:42 loss 1.0537 (1.0652) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.880, TIME@all 0.312 -epoch: [286/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:48 loss 1.0989 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [286/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:16:42 loss 1.0616 (1.0649) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.888, TIME@all 0.312 -epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0584 (1.0579) acc 100.0000 (99.8438) lr 0.002600 -epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0540 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 809.534, TIME@all 0.316 -epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.1104 (1.0629) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0869 (1.0720) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 809.587, TIME@all 0.316 -epoch: [287/350][20/50] time 0.327 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0730 (1.0558) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0636 (1.0688) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 809.411, TIME@all 0.316 -epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0537 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0583 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 809.433, TIME@all 0.316 -epoch: [287/350][20/50] time 0.326 (0.317) data 0.001 (0.012) eta 0:16:48 loss 1.0844 (1.0545) acc 96.8750 (99.8438) lr 0.002600 -epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:44 loss 1.0727 (1.0621) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 809.681, TIME@all 0.316 -epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.013) eta 0:16:48 loss 1.0773 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.309 (0.318) data 0.000 (0.007) eta 0:16:44 loss 1.1349 (1.0715) acc 96.8750 (99.6875) lr 0.002600 -FPS@all 809.476, TIME@all 0.316 -epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0592 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.309 (0.318) data 0.001 (0.006) eta 0:16:45 loss 1.0626 (1.0639) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 809.463, TIME@all 0.316 -epoch: [287/350][20/50] time 0.326 (0.317) data 0.000 (0.012) eta 0:16:48 loss 1.0545 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [287/350][40/50] time 0.310 (0.318) data 0.000 (0.006) eta 0:16:45 loss 1.0549 (1.0602) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 809.463, TIME@all 0.316 -epoch: [288/350][20/50] time 0.315 (0.312) data 0.001 (0.013) eta 0:16:18 loss 1.0604 (1.0586) acc 100.0000 (99.8438) lr 0.002600 -epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0494 (1.0633) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.770, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1134 (1.0587) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0448 (1.0679) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.545, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.001 (0.012) eta 0:16:18 loss 1.0775 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0492 (1.0636) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.584, TIME@all 0.312 -epoch: [288/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:16:18 loss 1.0500 (1.0567) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:16:13 loss 1.0692 (1.0651) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.379, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1051 (1.0560) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0424 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.397, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:16:18 loss 1.0759 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:16:13 loss 1.0611 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.419, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.1076 (1.0592) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0463 (1.0645) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.404, TIME@all 0.312 -epoch: [288/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:16:18 loss 1.0751 (1.0632) acc 100.0000 (100.0000) lr 0.002600 -epoch: [288/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:16:13 loss 1.0564 (1.0675) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.426, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.1081 (1.0575) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.0853 (1.0677) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.181, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.0888 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:15:56 loss 1.1359 (1.0701) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.214, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:02 loss 1.0776 (1.0581) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.1046 (1.0671) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.054, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.1520 (1.0639) acc 100.0000 (99.8438) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:15:56 loss 1.1033 (1.0726) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.091, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.0844 (1.0561) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:15:56 loss 1.1669 (1.0658) acc 96.8750 (99.8438) lr 0.002600 -FPS@all 820.205, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:16:01 loss 1.0908 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:15:56 loss 1.0621 (1.0648) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.166, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.0913 (1.0569) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:15:56 loss 1.0615 (1.0660) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.105, TIME@all 0.312 -epoch: [289/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:16:01 loss 1.1225 (1.0578) acc 100.0000 (100.0000) lr 0.002600 -epoch: [289/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:15:56 loss 1.0573 (1.0663) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.135, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0626 (1.0596) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0522 (1.0683) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.355, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0550 (1.0538) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0519 (1.0618) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 821.415, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.011) eta 0:15:46 loss 1.0574 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0579 (1.0686) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 821.261, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0723 (1.0554) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0806 (1.0708) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.242, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:45 loss 1.0434 (1.0553) acc 100.0000 (99.8438) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0502 (1.0694) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.399, TIME@all 0.312 -epoch: [290/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0447 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0469 (1.0670) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.293, TIME@all 0.312 -epoch: [290/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:15:46 loss 1.0732 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:15:39 loss 1.0470 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.264, TIME@all 0.312 -epoch: [290/350][20/50] time 0.316 (0.312) data 0.000 (0.012) eta 0:15:46 loss 1.0734 (1.0588) acc 100.0000 (100.0000) lr 0.002600 -epoch: [290/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:15:39 loss 1.0717 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 821.367, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0774 (1.0686) acc 100.0000 (99.6875) lr 0.002600 -epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.0667 (1.0695) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.563, TIME@all 0.312 -epoch: [291/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0631 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1014 (1.0654) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.489, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1826 (1.0606) acc 96.8750 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1119 (1.0726) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.444, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:15:31 loss 1.0806 (1.0544) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:15:25 loss 1.1424 (1.0671) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.457, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1142 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -epoch: [291/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.1173 (1.0681) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 820.501, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.1349 (1.0587) acc 96.8750 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.312 (0.313) data 0.001 (0.007) eta 0:15:25 loss 1.0753 (1.0643) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.491, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0536 (1.0608) acc 100.0000 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:15:25 loss 1.1294 (1.0729) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.457, TIME@all 0.312 -epoch: [291/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:15:31 loss 1.0645 (1.0566) acc 100.0000 (99.8438) lr 0.002600 -epoch: [291/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:15:25 loss 1.0740 (1.0668) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.480, TIME@all 0.312 -epoch: [292/350][20/50] time 0.318 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0825 (1.0589) acc 100.0000 (99.8438) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:15:12 loss 1.0515 (1.0698) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.191, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0834 (1.0598) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.1003 (1.0759) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.246, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.0626 (1.0502) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0442 (1.0636) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.112, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.001 (0.012) eta 0:15:15 loss 1.0747 (1.0539) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0721 (1.0679) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.122, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.001 (0.013) eta 0:15:15 loss 1.0679 (1.0541) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:15:12 loss 1.0752 (1.0695) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.133, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.013) eta 0:15:15 loss 1.0880 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.007) eta 0:15:12 loss 1.0434 (1.0685) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.137, TIME@all 0.313 -epoch: [292/350][20/50] time 0.317 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.0664 (1.0597) acc 100.0000 (100.0000) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:15:12 loss 1.0588 (1.0733) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.252, TIME@all 0.313 -epoch: [292/350][20/50] time 0.318 (0.312) data 0.000 (0.012) eta 0:15:15 loss 1.1226 (1.0619) acc 100.0000 (99.8438) lr 0.002600 -epoch: [292/350][40/50] time 0.312 (0.314) data 0.001 (0.006) eta 0:15:12 loss 1.0612 (1.0683) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.183, TIME@all 0.313 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:14:58 loss 1.0579 (1.0624) acc 100.0000 (99.8438) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.1036 (1.0694) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.797, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0569 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0873 (1.0709) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.707, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:14:59 loss 1.0576 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:14:53 loss 1.0811 (1.0706) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.633, TIME@all 0.312 -epoch: [293/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0972 (1.0589) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0651 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.586, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0472 (1.0552) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0590 (1.0662) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.654, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0889 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0953 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.735, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0517 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0549 (1.0701) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.676, TIME@all 0.312 -epoch: [293/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:14:59 loss 1.0607 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [293/350][40/50] time 0.309 (0.312) data 0.000 (0.007) eta 0:14:53 loss 1.0711 (1.0685) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.672, TIME@all 0.312 -epoch: [294/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0592 (1.0559) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:38 loss 1.0685 (1.0646) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.661, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0520 (1.0515) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:38 loss 1.0565 (1.0579) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.673, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.012) eta 0:14:47 loss 1.0537 (1.0576) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:14:39 loss 1.0503 (1.0661) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.583, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.001 (0.013) eta 0:14:47 loss 1.0724 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:14:39 loss 1.0793 (1.0610) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.586, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0788 (1.0562) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0484 (1.0635) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.540, TIME@all 0.313 -epoch: [294/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0433 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0612 (1.0677) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.632, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0665 (1.0534) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0652 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.584, TIME@all 0.313 -epoch: [294/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:14:47 loss 1.0542 (1.0586) acc 100.0000 (100.0000) lr 0.002600 -epoch: [294/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:14:39 loss 1.0868 (1.0674) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.590, TIME@all 0.313 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0479 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0747 (1.0691) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.536, TIME@all 0.312 -epoch: [295/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0520 (1.0600) acc 100.0000 (99.6875) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0884 (1.0719) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 820.393, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0569 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.1006 (1.0652) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.487, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0480 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.1213 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 820.456, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:14:30 loss 1.0518 (1.0571) acc 100.0000 (99.8438) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:14:23 loss 1.0800 (1.0664) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 820.444, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0525 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0976 (1.0705) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.493, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0457 (1.0553) acc 100.0000 (100.0000) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0644 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 820.544, TIME@all 0.312 -epoch: [295/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:14:30 loss 1.0457 (1.0598) acc 100.0000 (99.6875) lr 0.002600 -epoch: [295/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:14:23 loss 1.0938 (1.0703) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 820.446, TIME@all 0.312 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1829 (1.0609) acc 100.0000 (99.8438) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.001 (0.006) eta 0:14:03 loss 1.1275 (1.0726) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 824.548, TIME@all 0.310 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:14:10 loss 1.1264 (1.0622) acc 100.0000 (100.0000) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.313 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0700 (1.0699) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 824.579, TIME@all 0.310 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1059 (1.0600) acc 100.0000 (100.0000) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0415 (1.0675) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 824.471, TIME@all 0.311 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:11 loss 1.1242 (1.0593) acc 100.0000 (100.0000) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0494 (1.0634) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 824.453, TIME@all 0.311 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:11 loss 1.1269 (1.0628) acc 100.0000 (99.6875) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.006) eta 0:14:03 loss 1.0658 (1.0661) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 824.498, TIME@all 0.310 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:14:11 loss 1.1366 (1.0615) acc 96.8750 (99.8438) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0574 (1.0706) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 824.481, TIME@all 0.310 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.001 (0.013) eta 0:14:11 loss 1.1480 (1.0591) acc 100.0000 (100.0000) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.313 (0.311) data 0.000 (0.007) eta 0:14:03 loss 1.0515 (1.0649) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 824.512, TIME@all 0.310 -epoch: [296/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:14:10 loss 1.1488 (1.0603) acc 100.0000 (100.0000) lr 0.002600 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [296/350][40/50] time 0.312 (0.311) data 0.001 (0.006) eta 0:14:03 loss 1.0670 (1.0728) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 824.541, TIME@all 0.310 -epoch: [297/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0645 (1.0573) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0648 (1.0695) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.727, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0675 (1.0597) acc 100.0000 (99.8438) lr 0.002600 -epoch: [297/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0681 (1.0689) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.617, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0707 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0918 (1.0688) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 818.642, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.011) eta 0:13:59 loss 1.0804 (1.0524) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.1046 (1.0661) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 818.612, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.001 (0.013) eta 0:13:59 loss 1.0835 (1.0537) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:13:52 loss 1.0817 (1.0620) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.876, TIME@all 0.313 -epoch: [297/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0749 (1.0522) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:13:52 loss 1.1059 (1.0618) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 818.647, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:13:59 loss 1.0722 (1.0557) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0955 (1.0657) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.661, TIME@all 0.313 -epoch: [297/350][20/50] time 0.314 (0.313) data 0.001 (0.012) eta 0:13:59 loss 1.0673 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [297/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:13:52 loss 1.0995 (1.0660) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.700, TIME@all 0.313 -epoch: [298/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:13:41 loss 1.0528 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1882 (1.0699) acc 96.8750 (99.9219) lr 0.002600 -FPS@all 823.308, TIME@all 0.311 -epoch: [298/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.0604 (1.0649) acc 100.0000 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.1367 (1.0729) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 823.126, TIME@all 0.311 -epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:13:41 loss 1.0492 (1.0543) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.1494 (1.0747) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 823.154, TIME@all 0.311 -epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:13:41 loss 1.0649 (1.0604) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.0915 (1.0682) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 823.207, TIME@all 0.311 -epoch: [298/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:13:41 loss 1.0546 (1.0594) acc 100.0000 (100.0000) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1012 (1.0664) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 823.237, TIME@all 0.311 -epoch: [298/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:13:41 loss 1.0549 (1.0659) acc 100.0000 (99.5312) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:13:33 loss 1.1090 (1.0710) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 823.160, TIME@all 0.311 -epoch: [298/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.0591 (1.0607) acc 100.0000 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.0902 (1.0655) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 823.251, TIME@all 0.311 -epoch: [298/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:13:41 loss 1.1372 (1.0650) acc 96.8750 (99.8438) lr 0.002600 -epoch: [298/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:33 loss 1.0714 (1.0722) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 823.184, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:13:24 loss 1.0928 (1.0589) acc 100.0000 (99.8438) lr 0.002600 -epoch: [299/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1157 (1.0740) acc 96.8750 (99.7656) lr 0.002600 -FPS@all 822.569, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0796 (1.0548) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0625 (1.0675) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 822.372, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0635 (1.0545) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1252 (1.0700) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 822.412, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.1103 (1.0580) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0654 (1.0797) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 822.477, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.001 (0.012) eta 0:13:24 loss 1.1204 (1.0568) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.001 (0.006) eta 0:13:17 loss 1.0739 (1.0684) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 822.432, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.1289 (1.0565) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0868 (1.0724) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 822.389, TIME@all 0.311 -epoch: [299/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:13:24 loss 1.1288 (1.0655) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.0712 (1.0779) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 822.516, TIME@all 0.311 -epoch: [299/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:13:24 loss 1.0843 (1.0540) acc 100.0000 (100.0000) lr 0.002600 -epoch: [299/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:13:17 loss 1.1382 (1.0768) acc 100.0000 (99.7656) lr 0.002600 -FPS@all 822.469, TIME@all 0.311 -epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0596 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0628 (1.0666) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 819.108, TIME@all 0.313 -epoch: [300/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:13:13 loss 1.0819 (1.0583) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0673 (1.0696) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.147, TIME@all 0.313 -epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.012) eta 0:13:13 loss 1.0533 (1.0536) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.006) eta 0:13:06 loss 1.0753 (1.0651) acc 100.0000 (100.0000) lr 0.002600 -FPS@all 818.984, TIME@all 0.313 -epoch: [300/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:13:13 loss 1.0514 (1.0564) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.313) data 0.000 (0.007) eta 0:13:06 loss 1.0500 (1.0737) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.060, TIME@all 0.313 -epoch: [300/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0642 (1.0551) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:07 loss 1.0521 (1.0713) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 818.993, TIME@all 0.313 -epoch: [300/350][20/50] time 0.309 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0551 (1.0526) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:06 loss 1.0448 (1.0654) acc 100.0000 (99.9219) lr 0.002600 -FPS@all 819.045, TIME@all 0.313 -epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0566 (1.0594) acc 100.0000 (99.8438) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.314) data 0.001 (0.007) eta 0:13:06 loss 1.0637 (1.0733) acc 100.0000 (99.6875) lr 0.002600 -FPS@all 819.041, TIME@all 0.313 -epoch: [300/350][20/50] time 0.310 (0.314) data 0.000 (0.013) eta 0:13:13 loss 1.0602 (1.0585) acc 100.0000 (100.0000) lr 0.002600 -epoch: [300/350][40/50] time 0.318 (0.314) data 0.000 (0.007) eta 0:13:06 loss 1.0499 (1.0710) acc 100.0000 (99.8438) lr 0.002600 -FPS@all 819.070, TIME@all 0.313 -epoch: [301/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0735 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0515 (1.0663) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.410, TIME@all 0.312 -epoch: [301/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0663 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0766 (1.0658) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.209, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:54 loss 1.0454 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.007) eta 0:12:48 loss 1.0759 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.278, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0650 (1.0578) acc 100.0000 (99.8438) lr 0.000260 -epoch: [301/350][40/50] time 0.309 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0853 (1.0700) acc 100.0000 (99.4531) lr 0.000260 -FPS@all 821.316, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0501 (1.0526) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0651 (1.0620) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.325, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:12:54 loss 1.0628 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0666 (1.0658) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.276, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:12:54 loss 1.0619 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0666 (1.0620) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.234, TIME@all 0.312 -epoch: [301/350][20/50] time 0.310 (0.312) data 0.001 (0.012) eta 0:12:54 loss 1.0647 (1.0621) acc 100.0000 (99.6875) lr 0.000260 -epoch: [301/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:48 loss 1.0601 (1.0660) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.306, TIME@all 0.312 -epoch: [302/350][20/50] time 0.315 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0577 (1.0566) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0846 (1.0652) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.753, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0587 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.1220 (1.0652) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.642, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0575 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0737 (1.0657) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.597, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:12:38 loss 1.0518 (1.0610) acc 100.0000 (99.8438) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:12:33 loss 1.0776 (1.0651) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.688, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0631 (1.0578) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0722 (1.0650) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.719, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.001 (0.012) eta 0:12:38 loss 1.0615 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0926 (1.0676) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.641, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:12:38 loss 1.0720 (1.0524) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:33 loss 1.0742 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.655, TIME@all 0.312 -epoch: [302/350][20/50] time 0.314 (0.312) data 0.001 (0.012) eta 0:12:38 loss 1.0549 (1.0566) acc 100.0000 (100.0000) lr 0.000260 -epoch: [302/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:12:33 loss 1.0646 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.689, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:12:23 loss 1.0797 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0527 (1.0718) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.546, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0752 (1.0639) acc 100.0000 (99.6875) lr 0.000260 -epoch: [303/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:12:17 loss 1.1225 (1.0759) acc 96.8750 (99.6094) lr 0.000260 -FPS@all 820.381, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.011) eta 0:12:23 loss 1.0771 (1.0582) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0685 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.383, TIME@all 0.312 -epoch: [303/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0716 (1.0641) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.310 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.1099 (1.0734) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.453, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0505 (1.0559) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:12:17 loss 1.0893 (1.0697) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.262, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0973 (1.0572) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0614 (1.0683) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.419, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.1232 (1.0555) acc 100.0000 (100.0000) lr 0.000260 -epoch: [303/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0533 (1.0625) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.461, TIME@all 0.312 -epoch: [303/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:12:23 loss 1.0580 (1.0590) acc 100.0000 (99.8438) lr 0.000260 -epoch: [303/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:12:17 loss 1.0815 (1.0691) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.428, TIME@all 0.312 -epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0505 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.0822 (1.0649) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.150, TIME@all 0.312 -epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.014) eta 0:12:07 loss 1.0504 (1.0644) acc 100.0000 (99.6875) lr 0.000260 -epoch: [304/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.0865 (1.0748) acc 96.8750 (99.6875) lr 0.000260 -FPS@all 821.221, TIME@all 0.312 -epoch: [304/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0659 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0708 (1.0673) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.142, TIME@all 0.312 -epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0813 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.313) data 0.001 (0.007) eta 0:12:01 loss 1.0865 (1.0679) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.092, TIME@all 0.312 -epoch: [304/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:12:07 loss 1.0593 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0944 (1.0633) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.165, TIME@all 0.312 -epoch: [304/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:12:07 loss 1.0543 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0497 (1.0699) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.164, TIME@all 0.312 -epoch: [304/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:12:07 loss 1.0511 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.001 (0.007) eta 0:12:01 loss 1.0649 (1.0682) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.144, TIME@all 0.312 -epoch: [304/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:12:07 loss 1.0604 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [304/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:12:01 loss 1.1171 (1.0640) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.188, TIME@all 0.312 -epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:11:53 loss 1.0824 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.007) eta 0:11:47 loss 1.0551 (1.0734) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.893, TIME@all 0.312 -epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0724 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.1025 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.816, TIME@all 0.312 -epoch: [305/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0863 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0717 (1.0644) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.915, TIME@all 0.312 -epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0659 (1.0594) acc 100.0000 (99.8438) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0756 (1.0672) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.832, TIME@all 0.312 -epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.011) eta 0:11:53 loss 1.0754 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0507 (1.0671) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.825, TIME@all 0.312 -epoch: [305/350][20/50] time 0.310 (0.313) data 0.000 (0.012) eta 0:11:53 loss 1.0483 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0552 (1.0692) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.826, TIME@all 0.312 -epoch: [305/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:11:53 loss 1.1160 (1.0593) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.302 (0.313) data 0.000 (0.006) eta 0:11:46 loss 1.0649 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.008, TIME@all 0.312 -epoch: [305/350][20/50] time 0.310 (0.313) data 0.001 (0.012) eta 0:11:53 loss 1.0852 (1.0623) acc 100.0000 (100.0000) lr 0.000260 -epoch: [305/350][40/50] time 0.303 (0.313) data 0.000 (0.006) eta 0:11:47 loss 1.0530 (1.0658) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.917, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1155 (1.0596) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0760 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.137, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:11:37 loss 1.1228 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:11:30 loss 1.0664 (1.0676) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.165, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1138 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.1242 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.033, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1054 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0977 (1.0722) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.097, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1765 (1.0634) acc 96.8750 (99.8438) lr 0.000260 -epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0992 (1.0706) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.104, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.1045 (1.0596) acc 100.0000 (100.0000) lr 0.000260 -epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.1213 (1.0704) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.159, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:11:37 loss 1.0818 (1.0641) acc 100.0000 (99.8438) lr 0.000260 -epoch: [306/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:11:30 loss 1.1011 (1.0728) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.101, TIME@all 0.312 -epoch: [306/350][20/50] time 0.315 (0.313) data 0.000 (0.012) eta 0:11:37 loss 1.0861 (1.0599) acc 100.0000 (99.8438) lr 0.000260 -epoch: [306/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:11:30 loss 1.0746 (1.0645) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.111, TIME@all 0.312 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0864 (1.0634) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0470 (1.0727) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.204, TIME@all 0.312 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0759 (1.0602) acc 100.0000 (99.6875) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0666 (1.0655) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.120, TIME@all 0.313 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0709 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0582 (1.0714) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.054, TIME@all 0.313 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.1140 (1.0686) acc 100.0000 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0498 (1.0701) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.042, TIME@all 0.313 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0521 (1.0616) acc 100.0000 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:11:15 loss 1.0553 (1.0750) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.080, TIME@all 0.313 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:11:21 loss 1.0597 (1.0710) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:11:15 loss 1.0552 (1.0756) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.126, TIME@all 0.313 -epoch: [307/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0657 (1.0612) acc 100.0000 (99.8438) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:11:15 loss 1.0913 (1.0693) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.098, TIME@all 0.313 -epoch: [307/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:11:21 loss 1.0647 (1.0626) acc 100.0000 (100.0000) lr 0.000260 -epoch: [307/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:11:15 loss 1.0603 (1.0726) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.077, TIME@all 0.313 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.1280 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0420 (1.0651) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.278, TIME@all 0.312 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0790 (1.0513) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0677 (1.0709) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 820.186, TIME@all 0.312 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0863 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0551 (1.0666) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.083, TIME@all 0.312 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.1026 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:11:00 loss 1.0450 (1.0636) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.126, TIME@all 0.312 -epoch: [308/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0904 (1.0577) acc 100.0000 (99.8438) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0499 (1.0621) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.183, TIME@all 0.312 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.014) eta 0:11:08 loss 1.1121 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0491 (1.0678) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.120, TIME@all 0.312 -epoch: [308/350][20/50] time 0.312 (0.314) data 0.000 (0.013) eta 0:11:08 loss 1.0853 (1.0529) acc 100.0000 (100.0000) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0494 (1.0652) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.094, TIME@all 0.312 -epoch: [308/350][20/50] time 0.313 (0.314) data 0.000 (0.014) eta 0:11:08 loss 1.1473 (1.0644) acc 100.0000 (99.8438) lr 0.000260 -epoch: [308/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:11:00 loss 1.0503 (1.0691) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.145, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:49 loss 1.0724 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0588 (1.0642) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.509, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:10:50 loss 1.0706 (1.0668) acc 100.0000 (99.5312) lr 0.000260 -epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0682 (1.0702) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 821.526, TIME@all 0.312 -epoch: [309/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.1391 (1.0640) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0891 (1.0687) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.385, TIME@all 0.312 -epoch: [309/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0679 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.1503 (1.0690) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 821.369, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0596 (1.0570) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0582 (1.0652) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.415, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0512 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0585 (1.0675) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.476, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:10:50 loss 1.0604 (1.0597) acc 100.0000 (99.8438) lr 0.000260 -epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.0712 (1.0670) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.421, TIME@all 0.312 -epoch: [309/350][20/50] time 0.311 (0.313) data 0.000 (0.014) eta 0:10:50 loss 1.0495 (1.0596) acc 100.0000 (99.6875) lr 0.000260 -epoch: [309/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:10:43 loss 1.1042 (1.0658) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.451, TIME@all 0.312 -epoch: [310/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:10:33 loss 1.0779 (1.0541) acc 100.0000 (99.8438) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:10:29 loss 1.0790 (1.0633) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.340, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0481 (1.0544) acc 100.0000 (99.8438) lr 0.000260 -epoch: [310/350][40/50] time 0.318 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0616 (1.0676) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.247, TIME@all 0.313 -epoch: [310/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0453 (1.0556) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0697 (1.0666) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.150, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:34 loss 1.0552 (1.0510) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0784 (1.0625) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.160, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:10:33 loss 1.0762 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.007) eta 0:10:29 loss 1.0762 (1.0651) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.191, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0646 (1.0599) acc 100.0000 (99.8438) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0695 (1.0725) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 818.243, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0675 (1.0518) acc 100.0000 (100.0000) lr 0.000260 -epoch: [310/350][40/50] time 0.320 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.0455 (1.0652) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.235, TIME@all 0.313 -epoch: [310/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:10:33 loss 1.0703 (1.0585) acc 100.0000 (99.8438) lr 0.000260 -epoch: [310/350][40/50] time 0.319 (0.313) data 0.000 (0.006) eta 0:10:29 loss 1.1379 (1.0681) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.268, TIME@all 0.313 -epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0979 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0633 (1.0660) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.719, TIME@all 0.312 -epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0604 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0522 (1.0623) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.715, TIME@all 0.312 -epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0520 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0675 (1.0607) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.549, TIME@all 0.312 -epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0626 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:10:12 loss 1.0691 (1.0616) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.653, TIME@all 0.312 -epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:10:19 loss 1.0681 (1.0514) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:10:13 loss 1.0478 (1.0623) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.618, TIME@all 0.312 -epoch: [311/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0715 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0806 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.682, TIME@all 0.312 -epoch: [311/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:10:19 loss 1.0575 (1.0550) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0665 (1.0622) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.654, TIME@all 0.312 -epoch: [311/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:10:19 loss 1.0643 (1.0513) acc 100.0000 (100.0000) lr 0.000260 -epoch: [311/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:10:13 loss 1.0641 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.660, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:10:02 loss 1.0485 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1430 (1.0677) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.100, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0513 (1.0620) acc 100.0000 (99.8438) lr 0.000260 -epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1363 (1.0683) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.010, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0462 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1134 (1.0649) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.881, TIME@all 0.312 -epoch: [312/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0466 (1.0655) acc 100.0000 (99.8438) lr 0.000260 -epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.0600 (1.0667) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.940, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0671 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1235 (1.0665) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.928, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0773 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1089 (1.0651) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.009, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0575 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:09:56 loss 1.0661 (1.0683) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.966, TIME@all 0.312 -epoch: [312/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:10:02 loss 1.0500 (1.0601) acc 100.0000 (100.0000) lr 0.000260 -epoch: [312/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:09:56 loss 1.1494 (1.0668) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.969, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:09:46 loss 1.0860 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0842 (1.0659) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.439, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.1177 (1.0673) acc 100.0000 (99.8438) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0673 (1.0765) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.260, TIME@all 0.312 -epoch: [313/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0515 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0461 (1.0679) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.357, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0544 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0663 (1.0708) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.254, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.013) eta 0:09:46 loss 1.0644 (1.0635) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0719 (1.0689) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.289, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.001 (0.013) eta 0:09:46 loss 1.0646 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [313/350][40/50] time 0.323 (0.313) data 0.000 (0.007) eta 0:09:42 loss 1.0703 (1.0693) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.348, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0504 (1.0548) acc 100.0000 (99.8438) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0596 (1.0680) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.346, TIME@all 0.312 -epoch: [313/350][20/50] time 0.310 (0.312) data 0.000 (0.012) eta 0:09:46 loss 1.0630 (1.0584) acc 100.0000 (99.8438) lr 0.000260 -epoch: [313/350][40/50] time 0.322 (0.313) data 0.000 (0.006) eta 0:09:42 loss 1.0659 (1.0683) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.333, TIME@all 0.312 -epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0636 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0679 (1.0620) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.867, TIME@all 0.312 -epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0483 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0601 (1.0650) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.901, TIME@all 0.312 -epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0682 (1.0562) acc 100.0000 (99.8438) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:09:25 loss 1.1156 (1.0727) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.799, TIME@all 0.312 -epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.012) eta 0:09:31 loss 1.0514 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:09:25 loss 1.0639 (1.0623) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.791, TIME@all 0.312 -epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.012) eta 0:09:31 loss 1.0893 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:09:25 loss 1.1218 (1.0691) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.905, TIME@all 0.312 -epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0627 (1.0498) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:09:25 loss 1.0810 (1.0596) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.809, TIME@all 0.312 -epoch: [314/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0515 (1.0520) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 0:09:25 loss 1.0493 (1.0665) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.850, TIME@all 0.312 -epoch: [314/350][20/50] time 0.308 (0.312) data 0.000 (0.013) eta 0:09:31 loss 1.0516 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [314/350][40/50] time 0.315 (0.312) data 0.001 (0.006) eta 0:09:25 loss 1.0838 (1.0641) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.853, TIME@all 0.312 -epoch: [315/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0772 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0742 (1.0726) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.839, TIME@all 0.312 -epoch: [315/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0878 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0831 (1.0682) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.704, TIME@all 0.312 -epoch: [315/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0829 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1058 (1.0693) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.736, TIME@all 0.312 -epoch: [315/350][20/50] time 0.312 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0809 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1148 (1.0683) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 820.756, TIME@all 0.312 -epoch: [315/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.1072 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0846 (1.0648) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.781, TIME@all 0.312 -epoch: [315/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0903 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.0961 (1.0686) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.754, TIME@all 0.312 -epoch: [315/350][20/50] time 0.311 (0.312) data 0.001 (0.012) eta 0:09:15 loss 1.0733 (1.0618) acc 100.0000 (99.8438) lr 0.000260 -epoch: [315/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:09:09 loss 1.1059 (1.0675) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.770, TIME@all 0.312 -epoch: [315/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:09:15 loss 1.0968 (1.0587) acc 100.0000 (100.0000) lr 0.000260 -epoch: [315/350][40/50] time 0.312 (0.312) data 0.001 (0.006) eta 0:09:09 loss 1.0558 (1.0643) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.731, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0584 (1.0541) acc 100.0000 (99.8438) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0668 (1.0657) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.112, TIME@all 0.312 -epoch: [316/350][20/50] time 0.318 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0457 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0883 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.987, TIME@all 0.312 -epoch: [316/350][20/50] time 0.318 (0.313) data 0.000 (0.011) eta 0:09:00 loss 1.0502 (1.0547) acc 100.0000 (99.8438) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0728 (1.0677) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.939, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0558 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0515 (1.0652) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.959, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0478 (1.0609) acc 100.0000 (99.8438) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0542 (1.0673) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.978, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0498 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0813 (1.0634) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.997, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0530 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:08:54 loss 1.0561 (1.0652) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.989, TIME@all 0.312 -epoch: [316/350][20/50] time 0.317 (0.313) data 0.000 (0.012) eta 0:09:00 loss 1.0703 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [316/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:08:54 loss 1.0559 (1.0632) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.961, TIME@all 0.312 -epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:08:44 loss 1.0597 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.1243 (1.0699) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.947, TIME@all 0.311 -epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0520 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0640 (1.0704) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.890, TIME@all 0.311 -epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0557 (1.0674) acc 100.0000 (99.6875) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0852 (1.0742) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 821.803, TIME@all 0.312 -epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0793 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0482 (1.0656) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.816, TIME@all 0.312 -epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0589 (1.0631) acc 100.0000 (99.8438) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0560 (1.0685) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.839, TIME@all 0.311 -epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0551 (1.0654) acc 100.0000 (99.5312) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0795 (1.0732) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 821.870, TIME@all 0.311 -epoch: [317/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:08:44 loss 1.0610 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:08:37 loss 1.0693 (1.0673) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.863, TIME@all 0.311 -epoch: [317/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:08:44 loss 1.0489 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [317/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:08:37 loss 1.0641 (1.0631) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.810, TIME@all 0.312 -epoch: [318/350][20/50] time 0.315 (0.312) data 0.000 (0.014) eta 0:08:29 loss 1.0519 (1.0521) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:08:23 loss 1.0562 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.076, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0893 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0594 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.914, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:08:28 loss 1.0715 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:08:24 loss 1.0655 (1.0662) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.951, TIME@all 0.313 -epoch: [318/350][20/50] time 0.315 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0675 (1.0571) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0519 (1.0681) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.001, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0527 (1.0600) acc 100.0000 (99.8438) lr 0.000260 -epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0499 (1.0688) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.998, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0673 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:08:24 loss 1.0642 (1.0653) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.993, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:29 loss 1.0658 (1.0625) acc 100.0000 (99.8438) lr 0.000260 -epoch: [318/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0660 (1.0689) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.948, TIME@all 0.313 -epoch: [318/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:08:28 loss 1.0908 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [318/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:08:24 loss 1.0686 (1.0629) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.970, TIME@all 0.313 -epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:08:13 loss 1.0868 (1.0626) acc 100.0000 (99.6875) lr 0.000260 -epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:06 loss 1.0413 (1.0629) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.607, TIME@all 0.312 -epoch: [319/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:08:13 loss 1.1006 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:06 loss 1.1227 (1.0642) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.666, TIME@all 0.312 -epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.1053 (1.0622) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0565 (1.0641) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.512, TIME@all 0.312 -epoch: [319/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0490 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0758 (1.0616) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.508, TIME@all 0.312 -epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0946 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:08:06 loss 1.0506 (1.0592) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.621, TIME@all 0.312 -epoch: [319/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:08:13 loss 1.0734 (1.0563) acc 100.0000 (99.8438) lr 0.000260 -epoch: [319/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:08:07 loss 1.0532 (1.0624) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.534, TIME@all 0.312 -epoch: [319/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:08:13 loss 1.0852 (1.0605) acc 100.0000 (99.8438) lr 0.000260 -epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.006) eta 0:08:07 loss 1.0568 (1.0645) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.550, TIME@all 0.312 -epoch: [319/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:08:13 loss 1.0931 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [319/350][40/50] time 0.317 (0.312) data 0.000 (0.007) eta 0:08:07 loss 1.1334 (1.0636) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 821.572, TIME@all 0.312 -epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.1170 (1.0590) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0602 (1.0653) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.284, TIME@all 0.312 -epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0842 (1.0620) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0567 (1.0720) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.190, TIME@all 0.312 -epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1147 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0675 (1.0712) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.270, TIME@all 0.312 -epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0756 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0463 (1.0619) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.189, TIME@all 0.312 -epoch: [320/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0744 (1.0558) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0827 (1.0652) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.264, TIME@all 0.312 -epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:07:58 loss 1.0834 (1.0601) acc 100.0000 (99.8438) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:51 loss 1.0526 (1.0654) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.212, TIME@all 0.312 -epoch: [320/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1019 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0490 (1.0626) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.227, TIME@all 0.312 -epoch: [320/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:07:58 loss 1.1420 (1.0606) acc 100.0000 (100.0000) lr 0.000260 -epoch: [320/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:07:51 loss 1.0470 (1.0662) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.229, TIME@all 0.312 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0854 (1.0625) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0983 (1.0662) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 816.530, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.1093 (1.0596) acc 100.0000 (99.8438) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:38 loss 1.1638 (1.0695) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 816.554, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.0617 (1.0621) acc 100.0000 (99.6875) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.1430 (1.0706) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 816.384, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.0615 (1.0588) acc 100.0000 (99.6875) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.0731 (1.0624) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 816.390, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.012) eta 0:07:45 loss 1.1208 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.311 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.1226 (1.0667) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 816.429, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0975 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0956 (1.0642) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 816.449, TIME@all 0.314 -epoch: [321/350][20/50] time 0.312 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.1480 (1.0621) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.006) eta 0:07:39 loss 1.0572 (1.0644) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 816.398, TIME@all 0.314 -epoch: [321/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:07:45 loss 1.0600 (1.0571) acc 100.0000 (100.0000) lr 0.000260 -epoch: [321/350][40/50] time 0.312 (0.314) data 0.000 (0.007) eta 0:07:39 loss 1.0844 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 816.438, TIME@all 0.314 -epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0448 (1.0662) acc 100.0000 (99.5312) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1174 (1.0738) acc 100.0000 (99.5312) lr 0.000260 -FPS@all 821.680, TIME@all 0.312 -epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0871 (1.0613) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1242 (1.0696) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.687, TIME@all 0.312 -epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:07:25 loss 1.0671 (1.0583) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0889 (1.0670) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.542, TIME@all 0.312 -epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.011) eta 0:07:25 loss 1.0527 (1.0589) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0902 (1.0638) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.533, TIME@all 0.312 -epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0508 (1.0627) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0740 (1.0680) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.596, TIME@all 0.312 -epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0543 (1.0581) acc 100.0000 (100.0000) lr 0.000260 -epoch: [322/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1159 (1.0698) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.628, TIME@all 0.312 -epoch: [322/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0607 (1.0628) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.0796 (1.0675) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.591, TIME@all 0.312 -epoch: [322/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:25 loss 1.0733 (1.0571) acc 100.0000 (99.8438) lr 0.000260 -epoch: [322/350][40/50] time 0.313 (0.312) data 0.000 (0.006) eta 0:07:19 loss 1.1503 (1.0697) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 821.546, TIME@all 0.312 -epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0611 (1.0527) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0829 (1.0646) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.789, TIME@all 0.313 -epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0518 (1.0567) acc 100.0000 (99.8438) lr 0.000260 -epoch: [323/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0552 (1.0660) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.837, TIME@all 0.313 -epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0500 (1.0531) acc 100.0000 (99.8438) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0974 (1.0629) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.760, TIME@all 0.313 -epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:07:11 loss 1.0471 (1.0507) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:07:05 loss 1.0873 (1.0650) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.689, TIME@all 0.313 -epoch: [323/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0522 (1.0547) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.1042 (1.0711) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 818.723, TIME@all 0.313 -epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0554 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0987 (1.0635) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.753, TIME@all 0.313 -epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0510 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0716 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.763, TIME@all 0.313 -epoch: [323/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:07:11 loss 1.0539 (1.0526) acc 100.0000 (100.0000) lr 0.000260 -epoch: [323/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:07:05 loss 1.0941 (1.0694) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.813, TIME@all 0.313 -epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0604 (1.0663) acc 100.0000 (99.6875) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0680 (1.0680) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.543, TIME@all 0.312 -epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:56 loss 1.0995 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:06:49 loss 1.1361 (1.0661) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 819.591, TIME@all 0.312 -epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0583 (1.0523) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0467 (1.0574) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.438, TIME@all 0.312 -epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0448 (1.0602) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0475 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.461, TIME@all 0.312 -epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0492 (1.0494) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.313 (0.313) data 0.001 (0.006) eta 0:06:50 loss 1.0465 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.558, TIME@all 0.312 -epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0553 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.001 (0.006) eta 0:06:50 loss 1.0670 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.495, TIME@all 0.312 -epoch: [324/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:56 loss 1.0513 (1.0523) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:06:50 loss 1.0610 (1.0613) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.499, TIME@all 0.312 -epoch: [324/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:06:56 loss 1.0779 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [324/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:06:50 loss 1.0609 (1.0640) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.481, TIME@all 0.312 -epoch: [325/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0610 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1183 (1.0682) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.033, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0561 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.1638 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.924, TIME@all 0.312 -epoch: [325/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0597 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:06:33 loss 1.1179 (1.0667) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.825, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0674 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.0729 (1.0621) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.845, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:06:40 loss 1.0554 (1.0562) acc 100.0000 (99.8438) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:06:33 loss 1.1291 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.889, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0776 (1.0619) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1381 (1.0696) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.860, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0615 (1.0601) acc 100.0000 (100.0000) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:06:33 loss 1.1161 (1.0660) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 820.916, TIME@all 0.312 -epoch: [325/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:06:40 loss 1.0510 (1.0635) acc 100.0000 (99.6875) lr 0.000260 -epoch: [325/350][40/50] time 0.315 (0.312) data 0.001 (0.007) eta 0:06:33 loss 1.0934 (1.0702) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.923, TIME@all 0.312 -epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0458 (1.0557) acc 100.0000 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0561 (1.0667) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.178, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:06:25 loss 1.0541 (1.0600) acc 100.0000 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0544 (1.0726) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.057, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0546 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0634 (1.0613) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.018, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0544 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0444 (1.0614) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.059, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.012) eta 0:06:25 loss 1.0626 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:18 loss 1.0579 (1.0626) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.099, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0554 (1.0537) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.1215 (1.0656) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 819.094, TIME@all 0.313 -epoch: [326/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:06:25 loss 1.0572 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [326/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0893 (1.0638) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.118, TIME@all 0.313 -epoch: [326/350][20/50] time 0.316 (0.314) data 0.000 (0.013) eta 0:06:25 loss 1.0700 (1.0567) acc 96.8750 (99.8438) lr 0.000260 -epoch: [326/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:06:18 loss 1.0470 (1.0629) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.058, TIME@all 0.313 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0589 (1.0570) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.310 (0.313) data 0.001 (0.006) eta 0:06:02 loss 1.0690 (1.0664) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.360, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0705 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0547 (1.0645) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.323, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0699 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.1064 (1.0647) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.233, TIME@all 0.312 -epoch: [327/350][20/50] time 0.308 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0566 (1.0514) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0658 (1.0635) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.208, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.1057 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.0533 (1.0690) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.316, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.001 (0.012) eta 0:06:09 loss 1.0597 (1.0560) acc 100.0000 (99.8438) lr 0.000260 -epoch: [327/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:06:02 loss 1.1235 (1.0688) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.276, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.013) eta 0:06:09 loss 1.0656 (1.0578) acc 100.0000 (99.8438) lr 0.000260 -epoch: [327/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:06:02 loss 1.0623 (1.0645) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.273, TIME@all 0.312 -epoch: [327/350][20/50] time 0.309 (0.313) data 0.000 (0.012) eta 0:06:09 loss 1.0651 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [327/350][40/50] time 0.312 (0.313) data 0.001 (0.006) eta 0:06:02 loss 1.0673 (1.0695) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.290, TIME@all 0.312 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:05:52 loss 1.1806 (1.0669) acc 96.8750 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:45 loss 1.0713 (1.0734) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 822.428, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.1453 (1.0626) acc 96.8750 (99.6875) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.1033 (1.0695) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.355, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.1064 (1.0624) acc 96.8750 (99.6875) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.1597 (1.0768) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 822.256, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0622 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0946 (1.0683) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 822.244, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0704 (1.0591) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0963 (1.0717) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 822.323, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0818 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0842 (1.0701) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 822.281, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:05:52 loss 1.1262 (1.0642) acc 100.0000 (100.0000) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:46 loss 1.0842 (1.0732) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.265, TIME@all 0.311 -epoch: [328/350][20/50] time 0.311 (0.312) data 0.000 (0.012) eta 0:05:52 loss 1.0647 (1.0594) acc 100.0000 (99.8438) lr 0.000260 -epoch: [328/350][40/50] time 0.315 (0.312) data 0.000 (0.006) eta 0:05:46 loss 1.0976 (1.0763) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 822.301, TIME@all 0.311 -epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0639 (1.0536) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0512 (1.0629) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.017, TIME@all 0.312 -epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0660 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0536 (1.0639) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.076, TIME@all 0.312 -epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:37 loss 1.0541 (1.0517) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:05:31 loss 1.1735 (1.0650) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 819.966, TIME@all 0.312 -epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0909 (1.0609) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0493 (1.0655) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.997, TIME@all 0.312 -epoch: [329/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0819 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:05:31 loss 1.0581 (1.0651) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.928, TIME@all 0.312 -epoch: [329/350][20/50] time 0.310 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0839 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0811 (1.0692) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.951, TIME@all 0.312 -epoch: [329/350][20/50] time 0.310 (0.313) data 0.001 (0.013) eta 0:05:37 loss 1.0694 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:05:31 loss 1.0641 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.010, TIME@all 0.312 -epoch: [329/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:05:37 loss 1.0844 (1.0729) acc 100.0000 (99.6875) lr 0.000260 -epoch: [329/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:05:31 loss 1.0590 (1.0731) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.960, TIME@all 0.312 -epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0690 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0577 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 822.983, TIME@all 0.311 -epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0456 (1.0552) acc 100.0000 (99.8438) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0812 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 823.021, TIME@all 0.311 -epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0552 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:05:14 loss 1.0527 (1.0642) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.834, TIME@all 0.311 -epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0527 (1.0606) acc 100.0000 (99.6875) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0490 (1.0662) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.915, TIME@all 0.311 -epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.012) eta 0:05:20 loss 1.0518 (1.0652) acc 100.0000 (99.6875) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.312) data 0.000 (0.006) eta 0:05:14 loss 1.0466 (1.0740) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.871, TIME@all 0.311 -epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0708 (1.0582) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0711 (1.0650) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 822.925, TIME@all 0.311 -epoch: [330/350][20/50] time 0.311 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0640 (1.0571) acc 100.0000 (99.8438) lr 0.000260 -epoch: [330/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:05:14 loss 1.0490 (1.0661) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 822.855, TIME@all 0.311 -epoch: [330/350][20/50] time 0.310 (0.311) data 0.000 (0.013) eta 0:05:20 loss 1.0885 (1.0590) acc 100.0000 (100.0000) lr 0.000260 -epoch: [330/350][40/50] time 0.314 (0.311) data 0.000 (0.007) eta 0:05:14 loss 1.0502 (1.0647) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 822.911, TIME@all 0.311 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:06 loss 1.0901 (1.0592) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:05:00 loss 1.0593 (1.0689) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.835, TIME@all 0.312 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0799 (1.0572) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0614 (1.0716) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.621, TIME@all 0.312 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0605 (1.0625) acc 100.0000 (99.8438) lr 0.000260 -epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0643 (1.0702) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.605, TIME@all 0.312 -epoch: [331/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0677 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0596 (1.0654) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.682, TIME@all 0.312 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0611 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0695 (1.0672) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.669, TIME@all 0.312 -epoch: [331/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0610 (1.0577) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0794 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.704, TIME@all 0.312 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:05:07 loss 1.0635 (1.0594) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0559 (1.0680) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.691, TIME@all 0.312 -epoch: [331/350][20/50] time 0.311 (0.313) data 0.001 (0.012) eta 0:05:06 loss 1.0577 (1.0593) acc 100.0000 (100.0000) lr 0.000260 -epoch: [331/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:05:00 loss 1.0797 (1.0681) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.742, TIME@all 0.312 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:52 loss 1.0767 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0561 (1.0667) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 817.682, TIME@all 0.313 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:52 loss 1.0670 (1.0560) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0951 (1.0703) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 817.681, TIME@all 0.313 -epoch: [332/350][20/50] time 0.313 (0.315) data 0.000 (0.013) eta 0:04:53 loss 1.0755 (1.0585) acc 100.0000 (99.8438) lr 0.000260 -epoch: [332/350][40/50] time 0.314 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0602 (1.0713) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 817.509, TIME@all 0.313 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.013) eta 0:04:53 loss 1.1067 (1.0590) acc 96.8750 (99.6875) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0651 (1.0731) acc 100.0000 (99.6094) lr 0.000260 -FPS@all 817.530, TIME@all 0.313 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.001 (0.013) eta 0:04:53 loss 1.0764 (1.0609) acc 100.0000 (99.6875) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0436 (1.0721) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 817.567, TIME@all 0.313 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.001 (0.014) eta 0:04:53 loss 1.0564 (1.0550) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.001 (0.007) eta 0:04:45 loss 1.0537 (1.0659) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 817.609, TIME@all 0.313 -epoch: [332/350][20/50] time 0.314 (0.315) data 0.000 (0.014) eta 0:04:53 loss 1.0473 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0729 (1.0685) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 817.572, TIME@all 0.313 -epoch: [332/350][20/50] time 0.313 (0.315) data 0.001 (0.013) eta 0:04:52 loss 1.0546 (1.0581) acc 100.0000 (100.0000) lr 0.000260 -epoch: [332/350][40/50] time 0.313 (0.314) data 0.000 (0.007) eta 0:04:45 loss 1.0479 (1.0749) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 817.660, TIME@all 0.313 -epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0808 (1.0644) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0820 (1.0628) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.462, TIME@all 0.312 -epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0444 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0738 (1.0634) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.356, TIME@all 0.312 -epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0981 (1.0629) acc 100.0000 (99.8438) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0474 (1.0640) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.293, TIME@all 0.312 -epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0506 (1.0603) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0610 (1.0637) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.277, TIME@all 0.312 -epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0710 (1.0593) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0719 (1.0668) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.331, TIME@all 0.312 -epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0648 (1.0600) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0962 (1.0667) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.306, TIME@all 0.312 -epoch: [333/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0706 (1.0545) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0739 (1.0621) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.306, TIME@all 0.312 -epoch: [333/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:04:34 loss 1.0720 (1.0563) acc 100.0000 (100.0000) lr 0.000260 -epoch: [333/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:04:28 loss 1.0611 (1.0597) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.339, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.015) eta 0:04:18 loss 1.0576 (1.0613) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.314 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0926 (1.0671) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.447, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0466 (1.0629) acc 100.0000 (99.8438) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0799 (1.0666) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.303, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0460 (1.0664) acc 100.0000 (99.8438) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1131 (1.0696) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.385, TIME@all 0.312 -epoch: [334/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0556 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.001 (0.007) eta 0:04:12 loss 1.0693 (1.0684) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.346, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0471 (1.0575) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1053 (1.0660) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.299, TIME@all 0.312 -epoch: [334/350][20/50] time 0.312 (0.312) data 0.000 (0.014) eta 0:04:18 loss 1.0460 (1.0531) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.1003 (1.0647) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.407, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.001 (0.013) eta 0:04:18 loss 1.0660 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0776 (1.0688) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.418, TIME@all 0.312 -epoch: [334/350][20/50] time 0.311 (0.312) data 0.000 (0.013) eta 0:04:18 loss 1.0558 (1.0617) acc 100.0000 (99.8438) lr 0.000260 -epoch: [334/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:04:12 loss 1.0774 (1.0659) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.402, TIME@all 0.312 -epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0738 (1.0588) acc 100.0000 (99.8438) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0834 (1.0678) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.776, TIME@all 0.313 -epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0673 (1.0669) acc 100.0000 (99.5312) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.1098 (1.0707) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 818.806, TIME@all 0.313 -epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0981 (1.0584) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:03:58 loss 1.0846 (1.0683) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.696, TIME@all 0.313 -epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0697 (1.0614) acc 100.0000 (99.8438) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0875 (1.0659) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.781, TIME@all 0.313 -epoch: [335/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0612 (1.0542) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0718 (1.0622) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.601, TIME@all 0.313 -epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0800 (1.0572) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.001 (0.007) eta 0:03:58 loss 1.0933 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.672, TIME@all 0.313 -epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:04:04 loss 1.0637 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:03:58 loss 1.0974 (1.0626) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.660, TIME@all 0.313 -epoch: [335/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:04:04 loss 1.0492 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [335/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:58 loss 1.0725 (1.0663) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.720, TIME@all 0.313 -epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0675 (1.0554) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0862 (1.0644) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.947, TIME@all 0.312 -epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.014) eta 0:03:48 loss 1.1224 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:41 loss 1.1045 (1.0634) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.011, TIME@all 0.312 -epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:03:48 loss 1.0631 (1.0537) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:03:42 loss 1.0745 (1.0646) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.877, TIME@all 0.312 -epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0647 (1.0556) acc 100.0000 (99.8438) lr 0.000260 -epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.1083 (1.0696) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 819.906, TIME@all 0.312 -epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0872 (1.0611) acc 100.0000 (99.8438) lr 0.000260 -epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:03:42 loss 1.2024 (1.0672) acc 93.7500 (99.6094) lr 0.000260 -FPS@all 819.902, TIME@all 0.312 -epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0578 (1.0515) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0598 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.954, TIME@all 0.312 -epoch: [336/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.0572 (1.0512) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0502 (1.0617) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.935, TIME@all 0.312 -epoch: [336/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:03:48 loss 1.1493 (1.0569) acc 100.0000 (100.0000) lr 0.000260 -epoch: [336/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:03:42 loss 1.0642 (1.0606) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.914, TIME@all 0.312 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.014) eta 0:03:32 loss 1.0526 (1.0596) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:03:26 loss 1.1048 (1.0750) acc 96.8750 (99.6875) lr 0.000260 -FPS@all 818.500, TIME@all 0.313 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:33 loss 1.0581 (1.0597) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0587 (1.0698) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 818.280, TIME@all 0.313 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0566 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0645 (1.0647) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.366, TIME@all 0.313 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0682 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0883 (1.0675) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.393, TIME@all 0.313 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0616 (1.0612) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0630 (1.0659) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.356, TIME@all 0.313 -epoch: [337/350][20/50] time 0.319 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0615 (1.0643) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0774 (1.0688) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.332, TIME@all 0.313 -epoch: [337/350][20/50] time 0.320 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0844 (1.0719) acc 100.0000 (99.8438) lr 0.000260 -epoch: [337/350][40/50] time 0.314 (0.314) data 0.001 (0.007) eta 0:03:26 loss 1.0720 (1.0728) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.411, TIME@all 0.313 -epoch: [337/350][20/50] time 0.320 (0.313) data 0.000 (0.013) eta 0:03:32 loss 1.0528 (1.0613) acc 100.0000 (100.0000) lr 0.000260 -epoch: [337/350][40/50] time 0.315 (0.314) data 0.000 (0.007) eta 0:03:26 loss 1.0717 (1.0717) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 818.377, TIME@all 0.313 -epoch: [338/350][20/50] time 0.306 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.0634 (1.0590) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0681 (1.0675) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.059, TIME@all 0.312 -epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.014) eta 0:03:16 loss 1.0836 (1.0588) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0463 (1.0626) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.076, TIME@all 0.312 -epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.0915 (1.0621) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0640 (1.0672) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.970, TIME@all 0.312 -epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.1081 (1.0584) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0595 (1.0653) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.965, TIME@all 0.312 -epoch: [338/350][20/50] time 0.306 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.1365 (1.0584) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0602 (1.0670) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 821.017, TIME@all 0.312 -epoch: [338/350][20/50] time 0.307 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.1009 (1.0557) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0574 (1.0611) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.023, TIME@all 0.312 -epoch: [338/350][20/50] time 0.307 (0.312) data 0.001 (0.013) eta 0:03:16 loss 1.0650 (1.0558) acc 100.0000 (99.8438) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0723 (1.0613) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.038, TIME@all 0.312 -epoch: [338/350][20/50] time 0.306 (0.312) data 0.000 (0.013) eta 0:03:16 loss 1.0768 (1.0562) acc 100.0000 (100.0000) lr 0.000260 -epoch: [338/350][40/50] time 0.313 (0.312) data 0.000 (0.007) eta 0:03:10 loss 1.0490 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.057, TIME@all 0.312 -epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0695 (1.0540) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0503 (1.0667) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.803, TIME@all 0.312 -epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:03:01 loss 1.0574 (1.0586) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:02:55 loss 1.0478 (1.0653) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.829, TIME@all 0.312 -epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0536 (1.0552) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0509 (1.0687) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.708, TIME@all 0.312 -epoch: [339/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0436 (1.0520) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0496 (1.0667) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.695, TIME@all 0.312 -epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0506 (1.0496) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0740 (1.0582) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.779, TIME@all 0.312 -epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:03:01 loss 1.0737 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:02:55 loss 1.0481 (1.0637) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.770, TIME@all 0.312 -epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0549 (1.0555) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0655 (1.0649) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.727, TIME@all 0.312 -epoch: [339/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:03:01 loss 1.0588 (1.0549) acc 100.0000 (99.8438) lr 0.000260 -epoch: [339/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:02:55 loss 1.0989 (1.0604) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.756, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0911 (1.0619) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0847 (1.0669) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.889, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.314 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0565 (1.0569) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0529 (1.0607) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.886, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.0528 (1.0549) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0468 (1.0586) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.778, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.314 (0.312) data 0.000 (0.011) eta 0:02:45 loss 1.0969 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0581 (1.0606) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.747, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.0611 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0485 (1.0566) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.817, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.1009 (1.0590) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.1056 (1.0669) acc 96.8750 (99.8438) lr 0.000260 -FPS@all 820.815, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.313 (0.311) data 0.000 (0.012) eta 0:02:45 loss 1.0856 (1.0568) acc 100.0000 (99.8438) lr 0.000260 -epoch: [340/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0842 (1.0674) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.790, TIME@all 0.312 -Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 8192.0 -epoch: [340/350][20/50] time 0.313 (0.312) data 0.000 (0.012) eta 0:02:45 loss 1.1098 (1.0608) acc 100.0000 (100.0000) lr 0.000260 -epoch: [340/350][40/50] time 0.311 (0.312) data 0.000 (0.006) eta 0:02:39 loss 1.0582 (1.0648) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.790, TIME@all 0.312 -epoch: [341/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.0687 (1.0564) acc 100.0000 (99.8438) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0511 (1.0651) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 820.330, TIME@all 0.312 -epoch: [341/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.0648 (1.0599) acc 100.0000 (99.8438) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0552 (1.0670) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.375, TIME@all 0.312 -epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.0765 (1.0598) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:02:23 loss 1.1107 (1.0676) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.243, TIME@all 0.312 -epoch: [341/350][20/50] time 0.312 (0.313) data 0.000 (0.012) eta 0:02:30 loss 1.0735 (1.0574) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.000 (0.006) eta 0:02:23 loss 1.1124 (1.0686) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.244, TIME@all 0.312 -epoch: [341/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:02:30 loss 1.1659 (1.0639) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:02:23 loss 1.0693 (1.0720) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.264, TIME@all 0.312 -epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.014) eta 0:02:30 loss 1.0822 (1.0605) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:02:23 loss 1.1309 (1.0720) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.264, TIME@all 0.312 -epoch: [341/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.1143 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.006) eta 0:02:23 loss 1.1079 (1.0650) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.352, TIME@all 0.312 -epoch: [341/350][20/50] time 0.312 (0.313) data 0.001 (0.013) eta 0:02:30 loss 1.0578 (1.0539) acc 100.0000 (100.0000) lr 0.000260 -epoch: [341/350][40/50] time 0.311 (0.313) data 0.001 (0.007) eta 0:02:23 loss 1.0594 (1.0665) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.262, TIME@all 0.312 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0487 (1.0595) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0773 (1.0668) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 817.959, TIME@all 0.313 -epoch: [342/350][20/50] time 0.314 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0526 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0959 (1.0638) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 817.874, TIME@all 0.313 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.001 (0.013) eta 0:02:14 loss 1.0464 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:02:08 loss 1.1071 (1.0742) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 818.011, TIME@all 0.313 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0498 (1.0553) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.1374 (1.0679) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 817.886, TIME@all 0.313 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0598 (1.0528) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0675 (1.0671) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 817.976, TIME@all 0.313 -epoch: [342/350][20/50] time 0.314 (0.313) data 0.000 (0.013) eta 0:02:14 loss 1.0878 (1.0567) acc 96.8750 (99.8438) lr 0.000260 -epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:02:08 loss 1.1174 (1.0701) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 817.958, TIME@all 0.313 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.013) eta 0:02:14 loss 1.0700 (1.0610) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0915 (1.0741) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 817.880, TIME@all 0.313 -epoch: [342/350][20/50] time 0.313 (0.313) data 0.000 (0.012) eta 0:02:14 loss 1.0520 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [342/350][40/50] time 0.314 (0.313) data 0.000 (0.006) eta 0:02:08 loss 1.0741 (1.0664) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 817.922, TIME@all 0.313 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1790 (1.0651) acc 96.8750 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0628 (1.0657) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.257, TIME@all 0.312 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1094 (1.0618) acc 100.0000 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0718 (1.0662) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.236, TIME@all 0.312 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.0799 (1.0599) acc 100.0000 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0724 (1.0641) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.142, TIME@all 0.313 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1809 (1.0588) acc 96.8750 (99.8438) lr 0.000260 -epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0603 (1.0635) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.144, TIME@all 0.313 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.012) eta 0:01:59 loss 1.1862 (1.0580) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.006) eta 0:01:52 loss 1.0564 (1.0684) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.087, TIME@all 0.313 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1345 (1.0550) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0522 (1.0661) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.214, TIME@all 0.312 -epoch: [343/350][20/50] time 0.313 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.1598 (1.0633) acc 100.0000 (100.0000) lr 0.000260 -epoch: [343/350][40/50] time 0.311 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0675 (1.0717) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.151, TIME@all 0.313 -epoch: [343/350][20/50] time 0.314 (0.314) data 0.000 (0.013) eta 0:01:59 loss 1.2321 (1.0618) acc 93.7500 (99.6875) lr 0.000260 -epoch: [343/350][40/50] time 0.310 (0.313) data 0.000 (0.007) eta 0:01:52 loss 1.0636 (1.0682) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 819.145, TIME@all 0.313 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.014) eta 0:01:42 loss 1.0571 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0518 (1.0586) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.193, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0577 (1.0532) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0450 (1.0577) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.087, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0762 (1.0499) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0864 (1.0638) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.040, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0481 (1.0535) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0508 (1.0603) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.111, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.001 (0.013) eta 0:01:42 loss 1.0566 (1.0555) acc 100.0000 (99.8438) lr 0.000260 -epoch: [344/350][40/50] time 0.311 (0.312) data 0.001 (0.007) eta 0:01:36 loss 1.0479 (1.0633) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.077, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0702 (1.0506) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0544 (1.0640) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.041, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0508 (1.0510) acc 100.0000 (99.8438) lr 0.000260 -epoch: [344/350][40/50] time 0.311 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0637 (1.0595) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.104, TIME@all 0.312 -epoch: [344/350][20/50] time 0.309 (0.312) data 0.000 (0.013) eta 0:01:42 loss 1.0489 (1.0536) acc 100.0000 (100.0000) lr 0.000260 -epoch: [344/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:01:36 loss 1.0778 (1.0649) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.109, TIME@all 0.312 -epoch: [345/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0945 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.319 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0594 (1.0617) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.408, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0892 (1.0644) acc 100.0000 (99.8438) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.006) eta 0:01:21 loss 1.0581 (1.0691) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.269, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0969 (1.0561) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0672 (1.0636) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.347, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.012) eta 0:01:27 loss 1.0729 (1.0535) acc 100.0000 (99.8438) lr 0.000260 -epoch: [345/350][40/50] time 0.319 (0.312) data 0.000 (0.006) eta 0:01:21 loss 1.0493 (1.0609) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.274, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.1895 (1.0618) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.001 (0.007) eta 0:01:21 loss 1.0500 (1.0695) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.384, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.2562 (1.0651) acc 96.8750 (99.8438) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0717 (1.0693) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.317, TIME@all 0.312 -epoch: [345/350][20/50] time 0.314 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0719 (1.0557) acc 100.0000 (99.8438) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0494 (1.0636) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.303, TIME@all 0.312 -epoch: [345/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:01:27 loss 1.0835 (1.0543) acc 100.0000 (100.0000) lr 0.000260 -epoch: [345/350][40/50] time 0.318 (0.312) data 0.000 (0.007) eta 0:01:21 loss 1.0898 (1.0644) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.291, TIME@all 0.312 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0497 (1.0546) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1122 (1.0617) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.456, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0555 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1038 (1.0626) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.407, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0624 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.007) eta 0:01:05 loss 1.1729 (1.0616) acc 96.8750 (99.9219) lr 0.000260 -FPS@all 818.507, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0662 (1.0521) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1349 (1.0619) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 818.407, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0744 (1.0579) acc 100.0000 (99.8438) lr 0.000260 -epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.1691 (1.0664) acc 96.8750 (99.6875) lr 0.000260 -FPS@all 818.427, TIME@all 0.313 -epoch: [346/350][20/50] time 0.325 (0.315) data 0.000 (0.013) eta 0:01:12 loss 1.0577 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.0776 (1.0586) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.470, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0602 (1.0514) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.314 (0.314) data 0.000 (0.006) eta 0:01:05 loss 1.0892 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.432, TIME@all 0.313 -epoch: [346/350][20/50] time 0.323 (0.315) data 0.000 (0.012) eta 0:01:12 loss 1.0562 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -epoch: [346/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:01:05 loss 1.0712 (1.0609) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 818.443, TIME@all 0.313 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0501 (1.0604) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0852 (1.0771) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 819.651, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0512 (1.0530) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0831 (1.0665) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.729, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0468 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.312 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0587 (1.0660) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.535, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:00:56 loss 1.0526 (1.0564) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0668 (1.0681) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.577, TIME@all 0.312 -epoch: [347/350][20/50] time 0.312 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0584 (1.0544) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0684 (1.0679) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.619, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:56 loss 1.0508 (1.0565) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.1010 (1.0696) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.610, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:56 loss 1.0490 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.006) eta 0:00:50 loss 1.0832 (1.0663) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.600, TIME@all 0.312 -epoch: [347/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:56 loss 1.0497 (1.0541) acc 100.0000 (100.0000) lr 0.000260 -epoch: [347/350][40/50] time 0.313 (0.313) data 0.000 (0.007) eta 0:00:50 loss 1.0699 (1.0660) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.703, TIME@all 0.312 -epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.014) eta 0:00:40 loss 1.1071 (1.0551) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.314 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0751 (1.0655) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.177, TIME@all 0.313 -epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.0941 (1.0540) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0486 (1.0642) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 819.179, TIME@all 0.313 -epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1209 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0496 (1.0656) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.064, TIME@all 0.313 -epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.012) eta 0:00:40 loss 1.2092 (1.0601) acc 96.8750 (99.8438) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.006) eta 0:00:34 loss 1.0625 (1.0714) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.071, TIME@all 0.313 -epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1372 (1.0558) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.1040 (1.0651) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.151, TIME@all 0.313 -epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1040 (1.0513) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0579 (1.0619) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.155, TIME@all 0.313 -epoch: [348/350][20/50] time 0.316 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1470 (1.0585) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0736 (1.0645) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 819.121, TIME@all 0.313 -epoch: [348/350][20/50] time 0.315 (0.313) data 0.000 (0.013) eta 0:00:40 loss 1.1153 (1.0579) acc 100.0000 (100.0000) lr 0.000260 -epoch: [348/350][40/50] time 0.315 (0.313) data 0.000 (0.007) eta 0:00:34 loss 1.0554 (1.0665) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 819.111, TIME@all 0.313 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0508 (1.0583) acc 100.0000 (99.8438) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0724 (1.0641) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.596, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0682 (1.0525) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:00:18 loss 1.0900 (1.0678) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 820.516, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.012) eta 0:00:24 loss 1.0553 (1.0566) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.006) eta 0:00:18 loss 1.0561 (1.0629) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.522, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0553 (1.0573) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0709 (1.0694) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.641, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0815 (1.0576) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.315 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0537 (1.0666) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.580, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0457 (1.0597) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0429 (1.0667) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.530, TIME@all 0.312 -epoch: [349/350][20/50] time 0.313 (0.312) data 0.000 (0.013) eta 0:00:24 loss 1.0662 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0616 (1.0621) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.578, TIME@all 0.312 -epoch: [349/350][20/50] time 0.312 (0.312) data 0.001 (0.013) eta 0:00:24 loss 1.0489 (1.0568) acc 100.0000 (100.0000) lr 0.000260 -epoch: [349/350][40/50] time 0.316 (0.312) data 0.000 (0.007) eta 0:00:18 loss 1.0460 (1.0665) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.626, TIME@all 0.312 -epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0722 (1.0533) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0573 (1.0664) acc 100.0000 (99.9219) lr 0.000260 -FPS@all 821.065, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0715 (1.0545) acc 100.0000 (99.8438) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0659 (1.0664) acc 100.0000 (99.7656) lr 0.000260 -FPS@all 821.109, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.013) eta 0:00:09 loss 1.0912 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0591 (1.0624) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 820.978, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.000 (0.012) eta 0:00:09 loss 1.0810 (1.0548) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.006) eta 0:00:03 loss 1.0499 (1.0673) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.976, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0627 (1.0663) acc 100.0000 (99.8438) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0509 (1.0708) acc 100.0000 (99.6875) lr 0.000260 -FPS@all 821.015, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0629 (1.0567) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0840 (1.0677) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.036, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.1065 (1.0538) acc 100.0000 (100.0000) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.001 (0.007) eta 0:00:03 loss 1.0758 (1.0587) acc 100.0000 (100.0000) lr 0.000260 -FPS@all 821.022, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -epoch: [350/350][20/50] time 0.311 (0.313) data 0.001 (0.013) eta 0:00:09 loss 1.0700 (1.0588) acc 100.0000 (99.8438) lr 0.000260 -epoch: [350/350][40/50] time 0.312 (0.312) data 0.000 (0.007) eta 0:00:03 loss 1.0651 (1.0657) acc 100.0000 (99.8438) lr 0.000260 -FPS@all 820.982, TIME@all 0.312 -=> Final test -##### Evaluating market1501 (source) ##### -Extracting features from query set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 3368-by-512 matrix -Extracting features from gallery set ... -Done, obtained 15913-by-512 matrix -Speed: 0.0306 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.5% -CMC curve -Rank-1 : 92.3% -Rank-5 : 97.1% -Rank-10 : 98.1% -Rank-20 : 98.8% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:25 -FPS@all 819.137, TIME@all 0.313 -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0296 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.6% -CMC curve -Rank-1 : 92.3% -Rank-5 : 97.2% -Rank-10 : 98.2% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:33 -FPS@all 819.169, TIME@all 0.313 -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0321 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.6% -CMC curve -Rank-1 : 92.3% -Rank-5 : 97.2% -Rank-10 : 98.1% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:38 -FPS@all 819.180, TIME@all 0.313 -Done, obtained 15913-by-512 matrix -Speed: 0.0315 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.5% -CMC curve -Rank-1 : 92.1% -Rank-5 : 97.0% -Rank-10 : 98.1% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:38 -FPS@all 819.124, TIME@all 0.313 -Done, obtained 15913-by-512 matrix -Speed: 0.0304 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.5% -CMC curve -Rank-1 : 92.2% -Rank-5 : 97.2% -Rank-10 : 98.2% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:41 -FPS@all 819.160, TIME@all 0.313 -Done, obtained 15913-by-512 matrix -Speed: 0.0357 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.5% -CMC curve -Rank-1 : 92.3% -Rank-5 : 97.2% -Rank-10 : 98.2% -Rank-20 : 98.8% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:41 -FPS@all 819.219, TIME@all 0.312 -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -Done, obtained 15913-by-512 matrix -Speed: 0.0328 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.6% -CMC curve -Rank-1 : 92.2% -Rank-5 : 97.2% -Rank-10 : 98.2% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:42 -FPS@all 819.277, TIME@all 0.312 -Done, obtained 15913-by-512 matrix -Speed: 0.0315 sec/batch -Computing distance matrix with metric=euclidean ... -Computing CMC and mAP ... -** Results ** -mAP: 79.5% -CMC curve -Rank-1 : 92.3% -Rank-5 : 97.1% -Rank-10 : 98.0% -Rank-20 : 98.9% -Checkpoint saved to "log/osnet_x1_0_market1501_softmax/model/model.pth.tar-350" -Elapsed 1:47:44 -FPS@all 819.199, TIME@all 0.313 -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -[INFO] Float status is overflow! -[INFO] Float status is overflow! -[INFO] Float status is overflow! -THPModule_npu_shutdown success. -- Gitee From db1132226f535159599f4da66ba6ce04e0c2e1fe Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:47:51 +0000 Subject: [PATCH 30/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/fusion=5Fresult.json?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../classification/OSNet/fusion_result.json | 87617 ---------------- 1 file changed, 87617 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/fusion_result.json diff --git a/PyTorch/contrib/cv/classification/OSNet/fusion_result.json b/PyTorch/contrib/cv/classification/OSNet/fusion_result.json deleted file mode 100644 index 2f45d05310..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/fusion_result.json +++ /dev/null @@ -1,87617 +0,0 @@ -{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "0_0" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "100_100" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "101_101" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "102_102" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "103_103" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "104_104" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "105_105" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "106_106" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "107_107" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "108_108" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "109_109" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "10_10" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "110_110" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "111_111" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "112_112" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "113_113" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "114_114" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "115_115" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "116_116" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "117_117" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "118_118" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "119_119" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "11_11" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "120_120" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "121_121" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "122_122" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "123_123" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "124_124" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "125_125" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "126_126" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "127_127" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "128_128" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "129_129" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "12_12" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "130_130" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "131_131" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "132_132" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "133_133" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "9" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "134_134" -}{ - "graph_fusion": { - "ConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "SplitConvConcatFusionPass": { - "effect_times": "0", - "match_times": "3" - }, - "ZConcatDFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "135_135" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "136_136" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "137_137" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "138_138" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "139_139" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "13_13" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "140_140" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "141_141" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "142_142" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "143_143" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "144_144" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "145_145" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "146_146" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "147_147" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "148_148" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "149_149" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "14_14" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "150_150" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "151_151" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "152_152" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "153_153" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "154_154" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "155_155" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "156_156" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "157_157" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "158_158" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "159_159" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "15_15" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "160_160" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "161_161" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "162_162", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "163_163" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "164_164" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "165_165" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "166_166" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "167_167" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "168_168" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "169_169" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "16_16" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "170_170" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "171_171" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "172_172" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "173_173" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "174_174" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "175_175" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "176_176" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "177_177" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "178_178" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "179_179" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "17_17" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "180_180" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "181_181" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "182_182" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "183_183" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "184_184", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "185_185" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "186_186" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "187_187" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "188_188" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "189_189" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "18_18" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "190_190" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "191_191" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "192_192" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "193_193" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "194_194" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "195_195" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "196_196" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "197_197" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "198_198" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "199_199" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "19_19" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "1_1" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "200_200" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "201_201" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "202_202" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "203_203" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "204_204" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "205_205" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "206_206" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "207_207" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "208_208" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "209_209" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "20_20" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "210_210" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "211_211" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "212_212" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "213_213" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "214_214" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "215_215" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "216_216" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "217_217" -}{ - "graph_fusion": { - "LogSoftmaxFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "218_218" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "219_219" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "21_21" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "220_220" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "221_221" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "222_222" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "223_223" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "224_224" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "225_225" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "226_226" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "227_227" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "228_228" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "229_229" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "22_22" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "230_230" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "231_231" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "232_232" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "233_233" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "234_234" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "235_235" -}{ - "graph_fusion": { - "RealDiv2MulsFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "236_236" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "237_237" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "238_238" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "239_239" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "23_23" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "240_240" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "241_241" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "242_242" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "243_243" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "244_244" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "245_245" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "3" - } - }, - "session_and_graph_id": "246_246" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "247_247" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "248_248" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "249_249" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "24_24" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "250_250" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "251_251" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "252_252" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "253_253" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "254_254" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "255_255" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "256_256" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "257_257" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "258_258" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "259_259" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "25_25" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "260_260" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "261_261" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "262_262" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "263_263" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "264_264" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "265_265" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "266_266" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "267_267" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "268_268" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "269_269" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "26_26" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "270_270" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "271_271" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "272_272" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "273_273" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "274_274" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "275_275" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "276_276" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "277_277" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "278_278" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "279_279" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "27_27" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "280_280" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "281_281" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "282_282" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "283_283" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "284_284" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "285_285" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "286_286" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "287_287" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "288_288" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "289_289" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "28_28" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "290_290" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "291_291" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "292_292" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "293_293" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "294_294" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "295_295" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "296_296" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "297_297" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "298_298" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "299_299" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "29_29" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "2_2" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "300_300" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "301_301" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "302_302" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "303_303" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "304_304" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "305_305" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "306_306" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "307_307" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "308_308" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "309_309" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "30_30" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "310_310" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "311_311" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "312_312" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "313_313" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "314_314" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "315_315" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "316_316" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "317_317" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "318_318" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "319_319" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "31_31" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "320_320" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "321_321" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "322_322" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "323_323" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "324_324" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "325_325" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "326_326" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "327_327" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "328_328" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "329_329" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "32_32" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "330_330" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "331_331" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "332_332" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "333_333" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "334_334" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "335_335" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "336_336" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "337_337" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "338_338" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "339_339" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "33_33" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "340_340" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "341_341" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "342_342" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "343_343" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "344_344" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "345_345" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "346_346" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "347_347" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "348_348" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "349_349" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "34_34" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "350_350" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "351_351" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "352_352" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "353_353" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "354_354" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "355_355" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "356_356" -}{ - "graph_fusion": { - "AvgPoolV2GradFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "357_357" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "358_358" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "359_359" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "35_35" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "360_360" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "361_361" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "362_362" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "363_363" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "364_364" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "365_365" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "366_366" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "367_367" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "368_368" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "369_369" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "36_36" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "370_370" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "371_371" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "372_372" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "373_373" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "374_374" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "375_375" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "376_376" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "377_377" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "378_378" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "379_379" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "37_37" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "380_380" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "381_381" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "382_382" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "383_383" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "384_384" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "385_385" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "386_386" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "387_387" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "388_388" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "389_389" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "38_38" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "390_390" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "391_391" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "392_392" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "393_393" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - }, - "TransposedUpdateFusionPass": { - "effect_times": "2", - "match_times": "2" - } - }, - "session_and_graph_id": "394_394" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "395_395" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "396_396" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "397_397" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "398_398" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpInputDilationFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "399_399" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "39_39" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "3_3" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "400_400" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "401_401" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "402_402" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "403_403" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "404_404" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "405_405" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "Conv2DbpFilterMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "406_406" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "407_407" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "408_408" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "409_409" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "40_40" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "410_410" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "411_411" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "412_412" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "413_413" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "414_414" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "415_415" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "416_416" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "417_417" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "418_418" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "419_419" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "41_41" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "420_420" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "421_421" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "422_422" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "423_423" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "424_424" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "425_425" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "426_426" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "427_427" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "428_428" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "429_429" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "42_42" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "430_430" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "431_431" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "432_432" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "433_433" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "434_434" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "435_435" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "436_436" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "437_437" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "438_438" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "439_439" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "43_43" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "440_440" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "441_441" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "442_442" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "443_443" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "444_444" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "445_445" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "446_446" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "447_447" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "448_448" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "449_449" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "44_44" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "450_450" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "451_451" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "452_452" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "453_453" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "454_454" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "455_455" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "456_456" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "457_457" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "458_458" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "459_459" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "45_45" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "460_460" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "461_461" -}{ - "graph_fusion": { - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "462_462" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "463_463" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "464_464" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "465_465" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "466_466" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TopKFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "467_467" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "468_468" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "469_469" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "46_46" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "470_470" -}{ - "graph_fusion": { - "AReduceSumFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrReduceSumFusion": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "471_471" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "472_472" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "473_473" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "474_474" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "475_475" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "476_476" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "477_477" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "478_478" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "479_479" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "47_47" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "480_480" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "481_481" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "482_482" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "483_483" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "484_484" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "485_485" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "486_486" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "487_487" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "488_488" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "489_489" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "48_48" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "490_490" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "491_491" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "492_492" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "493_493" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "494_494" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "495_495" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "496_496" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "497_497" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "498_498" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "499_499" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "49_49" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "4_4" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "500_500" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "501_501" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "502_502" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "503_503" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "504_504" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "505_505" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "506_506" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "507_507" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "508_508" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "509_509" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "50_50" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "510_510" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "511_511" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "512_512" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "513_513" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "514_514" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "515_515" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "516_516" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "517_517" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "518_518" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "519_519" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "51_51" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "520_520" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "521_521" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "522_522" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "523_523" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "524_524" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "525_525" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "526_526" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "527_527" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "528_528" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "529_529" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "52_52" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "530_530" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "531_531" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "532_532" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "533_533" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "534_534" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "535_535" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "536_536" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "537_537" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "538_538" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "539_539" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "53_53" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "540_540" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "541_541" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "542_542" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "543_543" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "544_544" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "545_545" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "546_546" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "547_547" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "548_548" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "549_549" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "54_54" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "550_550" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "551_551" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "552_552" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "553_553" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "554_554" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "555_555" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "556_556" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "557_557" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "558_558" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "559_559" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "55_55" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "560_560" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "561_561" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "562_562" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "563_563" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "564_564" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "565_565" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "566_566" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "567_567" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "568_568" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "569_569" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "56_56" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "570_570" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "571_571" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "572_572" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "573_573" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "574_574" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "575_575" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "576_576", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "577_577" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "578_578" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "579_579" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "57_57" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "580_580" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "581_581" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "582_582" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "583_583" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "584_584" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "585_585" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "586_586" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "587_587" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "588_588" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "589_589" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "58_58" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "590_590" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "591_591" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "592_592" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "593_593" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "594_594" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "595_595" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "596_596", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "597_597" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "598_598" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "599_599" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "59_59" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "5_5" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "600_600" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "601_601" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "602_602" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "603_603" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "604_604" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "605_605" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "606_606" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "607_607" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "608_608" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "609_609" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "60_60" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "610_610" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "611_611" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "612_612" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "613_613" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "614_614" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "615_615" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "616_616" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "617_617" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "618_618" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "619_619" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "61_61" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "620_620" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "621_621" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "622_622" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "623_623" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "624_624" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "625_625" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "626_626" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "627_627" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "628_628" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "629_629" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "62_62" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "630_630" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "631_631" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "632_632" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "633_633" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "634_634" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "635_635" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "636_636" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "637_637" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "638_638" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "639_639" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "63_63" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "640_640" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "641_641" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "642_642" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "643_643" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "644_644" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "645_645" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "646_646" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "647_647", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "648_648" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "649_649" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "64_64" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "650_650" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "651_651" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "652_652" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "653_653" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "654_654" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "655_655" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "656_656" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "657_657" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "658_658" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "659_659" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "65_65" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "660_660" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "661_661" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "662_662" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "663_663" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "664_664" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "665_665" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "666_666" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "667_667", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "668_668" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "669_669" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "66_66" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "670_670" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "671_671" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "672_672" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "673_673" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "674_674" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "675_675" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "676_676" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "677_677" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "678_678" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "679_679" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "67_67" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "680_680" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "681_681" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "682_682" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "683_683" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "684_684" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "685_685" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "686_686" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "687_687" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "688_688" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "689_689" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "68_68" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "690_690" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "691_691" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "692_692" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "693_693" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "694_694" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "695_695" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "696_696" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "697_697" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "698_698" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "699_699" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "69_69" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "6_6" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "700_700" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "701_701" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "702_702" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "703_703" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "704_704" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "705_705" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "706_706" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "707_707" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "708_708" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "709_709" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "70_70" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "710_710" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "711_711" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "712_712" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "713_713" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "714_714" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "715_715" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "716_716" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "717_717" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "718_718", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "719_719" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "71_71" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "720_720" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "721_721" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "722_722" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "723_723" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "724_724" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "725_725" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "726_726" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "727_727" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "728_728" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "729_729" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "72_72" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "730_730" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "731_731" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "732_732" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "733_733" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "734_734" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "735_735" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "736_736" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "737_737" -}{ - "graph_fusion": { - "AvgPoolV2FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "738_738", - "ub_fusion": { - "TbeCommonRules0FusionPass": { - "effect_times": "1", - "match_times": "1" - } - } -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "739_739" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "73_73" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "740_740" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "741_741" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "742_742" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "TransposedUpdateFusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "743_743" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "744_744" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "745_745" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "746_746" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "747_747" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "748_748" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "749_749" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "74_74" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "750_750" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "751_751" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "752_752" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "753_753" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "754_754" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "755_755" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "756_756" -}{ - "graph_fusion": { - "ConvToFullyConnectionFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConvWeightCompressFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "757_757" -}{ - "graph_fusion": { - "AReduceMeanFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ConstToAttrPass": { - "effect_times": "1", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "758_758" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "759_759" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "75_75" -}{ - "graph_fusion": { - "MulSquareFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "760_760" -}{ - "graph_fusion": { - "BatchMatMulFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEAPREQUANTFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "FIXPIPEFUSIONPASS": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "761_761" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "762_762" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "763_763" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "2" - } - }, - "session_and_graph_id": "764_764" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "765_765" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "766_766" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "76_76" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "77_77" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "78_78" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "79_79" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "7_7" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "80_80" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "81_81" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "82_82" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "83_83" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "84_84" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "85_85" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "86_86" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "87_87" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "88_88" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "89_89" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "8_8" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "90_90" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "91_91" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "92_92" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "93_93" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "94_94" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "95_95" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "96_96" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "97_97" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "98_98" -}{ - "graph_fusion": { - "CastRemoveFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "ForceFp16CastFusionPass": { - "effect_times": "0", - "match_times": "1" - }, - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - } - }, - "session_and_graph_id": "99_99" -}{ - "graph_fusion": { - "RefreshInt64ToInt32FusionPass": { - "effect_times": "1", - "match_times": "1" - }, - "TransdataCastFusionPass": { - "effect_times": "0", - "match_times": "1" - } - }, - "session_and_graph_id": "9_9" -} \ No newline at end of file -- Gitee From 2f50776f516b9803debafe4819b2fe015fd34659 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E7=99=BD=E5=AE=87?= <11100105+bbbb23333@user.noreply.gitee.com> Date: Thu, 9 Jun 2022 10:49:03 +0000 Subject: [PATCH 31/31] =?UTF-8?q?=E5=88=A0=E9=99=A4=E6=96=87=E4=BB=B6=20Py?= =?UTF-8?q?Torch/contrib/cv/classification/OSNet/README.rst?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../cv/classification/OSNet/README.rst | 317 ------------------ 1 file changed, 317 deletions(-) delete mode 100644 PyTorch/contrib/cv/classification/OSNet/README.rst diff --git a/PyTorch/contrib/cv/classification/OSNet/README.rst b/PyTorch/contrib/cv/classification/OSNet/README.rst deleted file mode 100644 index 57be7a86ba..0000000000 --- a/PyTorch/contrib/cv/classification/OSNet/README.rst +++ /dev/null @@ -1,317 +0,0 @@ -Torchreid -=========== -Torchreid is a library for deep-learning person re-identification, written in `PyTorch `_ and developed for our ICCV'19 project, `Omni-Scale Feature Learning for Person Re-Identification `_. - -It features: - -- multi-GPU training -- support both image- and video-reid -- end-to-end training and evaluation -- incredibly easy preparation of reid datasets -- multi-dataset training -- cross-dataset evaluation -- standard protocol used by most research papers -- highly extensible (easy to add models, datasets, training methods, etc.) -- implementations of state-of-the-art deep reid models -- access to pretrained reid models -- advanced training techniques -- visualization tools (tensorboard, ranks, etc.) - - -Code: https://github.com/KaiyangZhou/deep-person-reid. - -Documentation: https://kaiyangzhou.github.io/deep-person-reid/. - -How-to instructions: https://kaiyangzhou.github.io/deep-person-reid/user_guide. - -Model zoo: https://kaiyangzhou.github.io/deep-person-reid/MODEL_ZOO. - -Tech report: https://arxiv.org/abs/1910.10093. - -You can find some research projects that are built on top of Torchreid `here `_. - - -What's new ------------- -- [Aug 2021] We have released the ImageNet-pretrained models of ``osnet_ain_x0_75``, ``osnet_ain_x0_5`` and ``osnet_ain_x0_25``. The pretraining setup follows `pycls `_. -- [Apr 2021] We have updated the appendix in the `TPAMI version of OSNet `_ to include results in the multi-source domain generalization setting. The trained models can be found in the `Model Zoo `_. -- [Apr 2021] We have added a script to automate the process of calculating average results over multiple splits. For more details please see ``tools/parse_test_res.py``. -- [Apr 2021] ``v1.4.0``: We added the person search dataset, `CUHK-SYSU `_. Please see the `documentation `_ regarding how to download the dataset (it contains cropped person images). -- [Apr 2021] All models in the model zoo have been moved to google drive. Please raise an issue if any model's performance is inconsistent with the numbers shown in the model zoo page (could be caused by wrong links). -- [Mar 2021] `OSNet `_ will appear in the TPAMI journal! Compared with the conference version, which focuses on discriminative feature learning using the omni-scale building block, this journal extension further considers generalizable feature learning by integrating `instance normalization layers `_ with the OSNet architecture. We hope this journal paper can motivate more future work to taclke the generalization issue in cross-dataset re-ID. -- [Mar 2021] Generalization across domains (datasets) in person re-ID is crucial in real-world applications, which is closely related to the topic of *domain generalization*. Interested in learning how the field of domain generalization has developed over the last decade? Check our recent survey in this topic at https://arxiv.org/abs/2103.02503, with coverage on the history, datasets, related problems, methodologies, potential directions, and so on (*methods designed for generalizable re-ID are also covered*!). -- [Feb 2021] ``v1.3.6`` Added `University-1652 `_, a new dataset for multi-view multi-source geo-localization (credit to `Zhedong Zheng `_). -- [Feb 2021] ``v1.3.5``: Now the `cython code `_ works on Windows (credit to `lablabla `_). -- [Jan 2021] Our recent work, `MixStyle `_ (mixing instance-level feature statistics of samples of different domains for improving domain generalization), has been accepted to ICLR'21. The code has been released at https://github.com/KaiyangZhou/mixstyle-release where the person re-ID part is based on Torchreid. -- [Jan 2021] A new evaluation metric called `mean Inverse Negative Penalty (mINP)` for person re-ID has been introduced in `Deep Learning for Person Re-identification: A Survey and Outlook (TPAMI 2021) `_. Their code can be accessed at ``_. -- [Aug 2020] ``v1.3.3``: Fixed bug in ``visrank`` (caused by not unpacking ``dsetid``). -- [Aug 2020] ``v1.3.2``: Added ``_junk_pids`` to ``grid`` and ``prid``. This avoids using mislabeled gallery images for training when setting ``combineall=True``. -- [Aug 2020] ``v1.3.0``: (1) Added ``dsetid`` to the existing 3-tuple data source, resulting in ``(impath, pid, camid, dsetid)``. This variable denotes the dataset ID and is useful when combining multiple datasets for training (as a dataset indicator). E.g., when combining ``market1501`` and ``cuhk03``, the former will be assigned ``dsetid=0`` while the latter will be assigned ``dsetid=1``. (2) Added ``RandomDatasetSampler``. Analogous to ``RandomDomainSampler``, ``RandomDatasetSampler`` samples a certain number of images (``batch_size // num_datasets``) from each of specified datasets (the amount is determined by ``num_datasets``). -- [Aug 2020] ``v1.2.6``: Added ``RandomDomainSampler`` (it samples ``num_cams`` cameras each with ``batch_size // num_cams`` images to form a mini-batch). -- [Jun 2020] ``v1.2.5``: (1) Dataloader's output from ``__getitem__`` has been changed from ``list`` to ``dict``. Previously, an element, e.g. image tensor, was fetched with ``imgs=data[0]``. Now it should be obtained by ``imgs=data['img']``. See this `commit `_ for detailed changes. (2) Added ``k_tfm`` as an option to image data loader, which allows data augmentation to be applied ``k_tfm`` times *independently* to an image. If ``k_tfm > 1``, ``imgs=data['img']`` returns a list with ``k_tfm`` image tensors. -- [May 2020] Added the person attribute recognition code used in `Omni-Scale Feature Learning for Person Re-Identification (ICCV'19) `_. See ``projects/attribute_recognition/``. -- [May 2020] ``v1.2.1``: Added a simple API for feature extraction (``torchreid/utils/feature_extractor.py``). See the `documentation `_ for the instruction. -- [Apr 2020] Code for reproducing the experiments of `deep mutual learning `_ in the `OSNet paper `__ (Supp. B) has been released at ``projects/DML``. -- [Apr 2020] Upgraded to ``v1.2.0``. The engine class has been made more model-agnostic to improve extensibility. See `Engine `_ and `ImageSoftmaxEngine `_ for more details. Credit to `Dassl.pytorch `_. -- [Dec 2019] Our `OSNet paper `_ has been updated, with additional experiments (in section B of the supplementary) showing some useful techniques for improving OSNet's performance in practice. -- [Nov 2019] ``ImageDataManager`` can load training data from target datasets by setting ``load_train_targets=True``, and the train-loader can be accessed with ``train_loader_t = datamanager.train_loader_t``. This feature is useful for domain adaptation research. - - -Installation ---------------- - -Make sure `conda `_ is installed. - - -.. code-block:: bash - - # cd to your preferred directory and clone this repo - git clone https://github.com/KaiyangZhou/deep-person-reid.git - - # create environment - cd deep-person-reid/ - conda create --name torchreid python=3.7 - conda activate torchreid - - # install dependencies - # make sure `which python` and `which pip` point to the correct path - pip install -r requirements.txt - - # install torch and torchvision (select the proper cuda version to suit your machine) - conda install pytorch torchvision cudatoolkit=9.0 -c pytorch - - # install torchreid (don't need to re-build it if you modify the source code) - python setup.py develop - - -Get started: 30 seconds to Torchreid -------------------------------------- -1. Import ``torchreid`` - -.. code-block:: python - - import torchreid - -2. Load data manager - -.. code-block:: python - - datamanager = torchreid.data.ImageDataManager( - root="reid-data", - sources="market1501", - targets="market1501", - height=256, - width=128, - batch_size_train=32, - batch_size_test=100, - transforms=["random_flip", "random_crop"] - ) - -3 Build model, optimizer and lr_scheduler - -.. code-block:: python - - model = torchreid.models.build_model( - name="resnet50", - num_classes=datamanager.num_train_pids, - loss="softmax", - pretrained=True - ) - - model = model.cuda() - - optimizer = torchreid.optim.build_optimizer( - model, - optim="adam", - lr=0.0003 - ) - - scheduler = torchreid.optim.build_lr_scheduler( - optimizer, - lr_scheduler="single_step", - stepsize=20 - ) - -4. Build engine - -.. code-block:: python - - engine = torchreid.engine.ImageSoftmaxEngine( - datamanager, - model, - optimizer=optimizer, - scheduler=scheduler, - label_smooth=True - ) - -5. Run training and test - -.. code-block:: python - - engine.run( - save_dir="log/resnet50", - max_epoch=60, - eval_freq=10, - print_freq=10, - test_only=False - ) - - -A unified interface ------------------------ -In "deep-person-reid/scripts/", we provide a unified interface to train and test a model. See "scripts/main.py" and "scripts/default_config.py" for more details. The folder "configs/" contains some predefined configs which you can use as a starting point. - -Below we provide an example to train and test `OSNet (Zhou et al. ICCV'19) `_. Assume :code:`PATH_TO_DATA` is the directory containing reid datasets. The environmental variable :code:`CUDA_VISIBLE_DEVICES` is omitted, which you need to specify if you have a pool of gpus and want to use a specific set of them. - -Conventional setting -^^^^^^^^^^^^^^^^^^^^^ - -To train OSNet on Market1501, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - --transforms random_flip random_erase \ - --root $PATH_TO_DATA - - -The config file sets Market1501 as the default dataset. If you wanna use DukeMTMC-reID, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - -s dukemtmcreid \ - -t dukemtmcreid \ - --transforms random_flip random_erase \ - --root $PATH_TO_DATA \ - data.save_dir log/osnet_x1_0_dukemtmcreid_softmax_cosinelr - -The code will automatically (download and) load the ImageNet pretrained weights. After the training is done, the model will be saved as "log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250". Under the same folder, you can find the `tensorboard `_ file. To visualize the learning curves using tensorboard, you can run :code:`tensorboard --logdir=log/osnet_x1_0_market1501_softmax_cosinelr` in the terminal and visit :code:`http://localhost:6006/` in your web browser. - -Evaluation is automatically performed at the end of training. To run the test again using the trained model, do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad_cosine.yaml \ - --root $PATH_TO_DATA \ - model.load_weights log/osnet_x1_0_market1501_softmax_cosinelr/model.pth.tar-250 \ - test.evaluate True - - -Cross-domain setting -^^^^^^^^^^^^^^^^^^^^^ - -Suppose you wanna train OSNet on DukeMTMC-reID and test its performance on Market1501, you can do - -.. code-block:: bash - - python scripts/main.py \ - --config-file configs/im_osnet_x1_0_softmax_256x128_amsgrad.yaml \ - -s dukemtmcreid \ - -t market1501 \ - --transforms random_flip color_jitter \ - --root $PATH_TO_DATA - -Here we only test the cross-domain performance. However, if you also want to test the performance on the source dataset, i.e. DukeMTMC-reID, you can set :code:`-t dukemtmcreid market1501`, which will evaluate the model on the two datasets separately. - -Different from the same-domain setting, here we replace :code:`random_erase` with :code:`color_jitter`. This can improve the generalization performance on the unseen target dataset. - -Pretrained models are available in the `Model Zoo `_. - - -Datasets --------- - -Image-reid datasets -^^^^^^^^^^^^^^^^^^^^^ -- `Market1501 `_ -- `CUHK03 `_ -- `DukeMTMC-reID `_ -- `MSMT17 `_ -- `VIPeR `_ -- `GRID `_ -- `CUHK01 `_ -- `SenseReID `_ -- `QMUL-iLIDS `_ -- `PRID `_ - -Geo-localization datasets -^^^^^^^^^^^^^^^^^^^^^^^^^^^ -- `University-1652 `_ - -Video-reid datasets -^^^^^^^^^^^^^^^^^^^^^^^ -- `MARS `_ -- `iLIDS-VID `_ -- `PRID2011 `_ -- `DukeMTMC-VideoReID `_ - - -Models -------- - -ImageNet classification models -^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -- `ResNet `_ -- `ResNeXt `_ -- `SENet `_ -- `DenseNet `_ -- `Inception-ResNet-V2 `_ -- `Inception-V4 `_ -- `Xception `_ -- `IBN-Net `_ - -Lightweight models -^^^^^^^^^^^^^^^^^^^ -- `NASNet `_ -- `MobileNetV2 `_ -- `ShuffleNet `_ -- `ShuffleNetV2 `_ -- `SqueezeNet `_ - -ReID-specific models -^^^^^^^^^^^^^^^^^^^^^^ -- `MuDeep `_ -- `ResNet-mid `_ -- `HACNN `_ -- `PCB `_ -- `MLFN `_ -- `OSNet `_ -- `OSNet-AIN `_ - - -Useful links -------------- -- `OSNet-IBN1-Lite (test-only code with lite docker container) `_ -- `Deep Learning for Person Re-identification: A Survey and Outlook `_ - - -Citation ---------- -If you use this code or the models in your research, please give credit to the following papers: - -.. code-block:: bash - - @article{torchreid, - title={Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch}, - author={Zhou, Kaiyang and Xiang, Tao}, - journal={arXiv preprint arXiv:1910.10093}, - year={2019} - } - - @inproceedings{zhou2019osnet, - title={Omni-Scale Feature Learning for Person Re-Identification}, - author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, - booktitle={ICCV}, - year={2019} - } - - @article{zhou2021osnet, - title={Learning Generalisable Omni-Scale Representations for Person Re-Identification}, - author={Zhou, Kaiyang and Yang, Yongxin and Cavallaro, Andrea and Xiang, Tao}, - journal={TPAMI}, - year={2021} - } -- Gitee