Commits · 8ea058b9bd31d16d8ae28f062d60acdc5af32150 · ModelZoo / ResNet50_tensorflow

28 May, 2020 1 commit
- Internal change · 980b27d5
  Abdullah Rashwan authored May 28, 2020
```
PiperOrigin-RevId: 313662797
```
  980b27d5
22 May, 2020 1 commit
- Internal change · fffcd235
  Yeqing Li authored May 21, 2020
```
PiperOrigin-RevId: 312770722
```
  fffcd235
07 May, 2020 1 commit
- Support constant argument in restriction checking in params_dict. · e9e6d17c
  A. Unique TensorFlower authored May 07, 2020
```
PiperOrigin-RevId: 310431197
```
  e9e6d17c
04 May, 2020 1 commit
- Classifier trainer GPU throughput fixes. · e91779b8
  Allen Wang authored May 04, 2020
```
PiperOrigin-RevId: 309776126
```
  e91779b8
01 May, 2020 2 commits
- Internal change · 71d2680d
  Yeqing Li authored May 01, 2020
```
PiperOrigin-RevId: 309486898
```
  71d2680d
- Export symbols of hyperparams module together. · 4eda0048
  Hongkun Yu authored May 01, 2020
```
PiperOrigin-RevId: 309457250
```
  4eda0048
21 Apr, 2020 1 commit
- ParamsDict update. · 9c1887a8
  Pengchong Jin authored Apr 21, 2020
```
PiperOrigin-RevId: 307639416
```
  9c1887a8
16 Apr, 2020 1 commit
- Support multiple metrics in CTL. · 5741cef6
  Yeqing Li authored Apr 15, 2020
```
PiperOrigin-RevId: 306751755
```
  5741cef6
14 Apr, 2020 1 commit

Set dataset private threads and GPU private threads together for better Resnet... · 0672bdcf

Zongwei Zhou authored Apr 13, 2020

Set dataset private threads and GPU private threads together for better Resnet performance; remove outdated TF1 benchmarks; remove redundant classifier param runtime.gpu_threads_enabled

PiperOrigin-RevId: 306341022

0672bdcf

10 Apr, 2020 1 commit
- Save the checkpoint at step 0 and run evaluation. · d10d0027
  Pengchong Jin authored Apr 09, 2020
```
PiperOrigin-RevId: 305806096
```
  d10d0027
03 Apr, 2020 1 commit
- Internal change · d3320242
  Will Cromar authored Apr 03, 2020
```
PiperOrigin-RevId: 304643558
```
  d3320242
01 Apr, 2020 1 commit
- Internal change · f276d472
  Allen Wang authored Apr 01, 2020
```
PiperOrigin-RevId: 304242451
```
  f276d472
26 Mar, 2020 1 commit
- Move modeling_training_utils.py to bert folder · 6ef71d49
  Hongkun Yu authored Mar 25, 2020
```
PiperOrigin-RevId: 303050809
```
  6ef71d49
25 Mar, 2020 1 commit

Check if there are GPUs instead of if TF is built with CUDA support. · ad09cf49

Reed Wanderman-Milne authored Mar 25, 2020

The TF pip packages are always built with CUDA support, so tf.test.is_built_with_cuda() would return True even if the user had no GPU.

PiperOrigin-RevId: 302928378

ad09cf49

17 Mar, 2020 3 commits
- Adds scale_loss to run_customized_training_loop, which should be the correct treatment. · bf6d6f6f
  Hongkun Yu authored Mar 17, 2020
```
PiperOrigin-RevId: 301441181
```
  bf6d6f6f
- tf.compat.v1.logging implemented with absl · 3043566d
  ayushmankumar7 authored Mar 18, 2020
  
  3043566d
- Internal change · 2d7223ad
  Ken Franko authored Mar 16, 2020
```
PiperOrigin-RevId: 301277545
```
  2d7223ad
15 Mar, 2020 2 commits
- Correct grammar. · adc01cd7
  A. Unique TensorFlower authored Mar 15, 2020
```
PiperOrigin-RevId: 301041987
```
  adc01cd7
- Internal change · 77e28369
  Hongkun Yu authored Mar 15, 2020
```
PiperOrigin-RevId: 301041141
```
  77e28369
13 Mar, 2020 1 commit
- Remove confusing legacy comments. · 5a423a2d
  Hongkun Yu authored Mar 12, 2020
```
PiperOrigin-RevId: 300693870
```
  5a423a2d
11 Mar, 2020 2 commits
- Add TimeHistory callback to BERT. · b39958cf
  Will Cromar authored Mar 11, 2020
```
PiperOrigin-RevId: 300433601
```
  b39958cf
- Internal change · 1792fb76
  Yeqing Li authored Mar 11, 2020
```
PiperOrigin-RevId: 300426073
```
  1792fb76
10 Mar, 2020 1 commit

Save to tmp directory on non-chief workers in model_training_utils · 682d36b5

Ran Chen authored Mar 10, 2020

In a multi worker set up saving is done on each worker. If they're saving to the same location, e.g. GCS, there will be conflicts. With this change we save to temporary directory on non-chief workers.

Note that, there may be synchronization in saving that needs all workers to participate, so we cannot only save on one worker.

PiperOrigin-RevId: 300141152

682d36b5

07 Mar, 2020 1 commit
- Add TimeHistory callback to BERT. · 7d86c317
  Hongkun Yu authored Mar 07, 2020
```
PiperOrigin-RevId: 299594839
```
  7d86c317
06 Mar, 2020 1 commit

Temporarily disable explicit allreduce in BERT SQuAD · 11ccb99e

Zongwei Zhou authored Mar 05, 2020

In BERT SQuAD, disable explicit allreduce for now to keep the original clip_by_global_norm math. With explicit allreduce, the gradients before allreduce are scaled so even if we move clip_by_global_norm before allreduce (as in TF1 and pre-TF 2.2) it will operate on scaled gradients, the math will be changed. So with explicit allreduce, it is better to move clip_by_global_norm to after allreduce.

PiperOrigin-RevId: 299278082

11ccb99e

05 Mar, 2020 1 commit
- Internal change · cf01596c
  Zongwei Zhou authored Mar 04, 2020
```
PiperOrigin-RevId: 299007295
```
  cf01596c
04 Mar, 2020 1 commit
- Remove debugging log. · b0ba8502
  Hongkun Yu authored Mar 03, 2020
```
PiperOrigin-RevId: 298717616
```
  b0ba8502
03 Mar, 2020 1 commit
- Moves some common config definition to modeliing/hyperparams/. · 1ac65814
  Yeqing Li authored Mar 03, 2020
```
PiperOrigin-RevId: 298652484
```
  1ac65814
02 Mar, 2020 2 commits
- Add TimeHistory callback to BERT. · 533d1e6b
  Will Cromar authored Mar 02, 2020
```
PiperOrigin-RevId: 298466825
```
  533d1e6b
- Remove an assert. · 98abe4b8
  Hongkun Yu authored Mar 02, 2020
```
PiperOrigin-RevId: 298402269
```
  98abe4b8
28 Feb, 2020 1 commit
- base_config.py in modeling/ · e1799db4
  Yeqing Li authored Feb 27, 2020
```
PiperOrigin-RevId: 297757527
```
  e1799db4
26 Feb, 2020 1 commit
- Internal change · ce83a9db
  Hongkun Yu authored Feb 26, 2020
```
PiperOrigin-RevId: 297383836
```
  ce83a9db
25 Feb, 2020 2 commits
- Internal change · 2781377d
  Zongwei Zhou authored Feb 25, 2020
```
PiperOrigin-RevId: 297222995
```
  2781377d
- Creates modeling/performance.py to include mix prediction related stuff · fb35d6be
  Hongkun Yu authored Feb 24, 2020
```
PiperOrigin-RevId: 297002741
```
  fb35d6be
20 Feb, 2020 1 commit
- Remove some tpu_lib not used at all · 58df0dc6
  Hongkun Yu authored Feb 19, 2020
```
PiperOrigin-RevId: 296140699
```
  58df0dc6
19 Feb, 2020 2 commits
- Use host python training loop in GPU BERT tests · 1ca9e3e4
  Zongwei Zhou authored Feb 18, 2020
```
PiperOrigin-RevId: 295869937
```
  1ca9e3e4
- Internal change · acea25b9
  A. Unique TensorFlower authored Feb 18, 2020
```
PiperOrigin-RevId: 295849975
```
  acea25b9
13 Feb, 2020 1 commit
- Revert log passing change since it might hurt performance. · 91c681af
  Haitang Hu authored Feb 13, 2020
```
PiperOrigin-RevId: 294922828
```
  91c681af
12 Feb, 2020 1 commit
- base_config.py in modeling/ · 6d1d918f
  Hongkun Yu authored Feb 11, 2020
```
PiperOrigin-RevId: 294560916
```
  6d1d918f
07 Feb, 2020 1 commit

Pass training_loss in logs dict to customized training loop. · 67f6015a

Haitang Hu authored Feb 07, 2020

This would match the behavior described in on_batch_end() functions in Keras callback.
See https://www.tensorflow.org/api_docs/python/tf/keras/callbacks/Callback.

PiperOrigin-RevId: 293887583

67f6015a