Commits · 0c80349880e826f59da070bb448af0711a45b991 · ModelZoo / ResNet50_tensorflow

26 Apr, 2021 2 commits
- Internal change · 0c803498
  Chen Chen authored Apr 26, 2021
```
PiperOrigin-RevId: 370538762
```
  0c803498
- Link to vision/beta/MODEL_GARDEN.md for resnet-rs link. · 6f1ddc42
  Hongkun Yu authored Apr 25, 2021
```
PiperOrigin-RevId: 370389563
```
  6f1ddc42
24 Apr, 2021 1 commit
- Internal change · 14cc4985
  Yeqing Li authored Apr 23, 2021
```
PiperOrigin-RevId: 370198074
```
  14cc4985
23 Apr, 2021 2 commits
- Internal change · 68ee72ae
  Chen Chen authored Apr 23, 2021
```
PiperOrigin-RevId: 370159521
```
  68ee72ae
- Internal change · 84b25763
  Jeremiah Liu authored Apr 23, 2021
```
PiperOrigin-RevId: 370096487
```
  84b25763
22 Apr, 2021 2 commits
- Internal change · d9ed5232
  Yeqing Li authored Apr 22, 2021
```
PiperOrigin-RevId: 369935333
```
  d9ed5232
- Convert model_garden/vision code to use public tf.keras API. · d54edbcf
  Scott Zhu authored Apr 22, 2021
```
PiperOrigin-RevId: 369893606
```
  d54edbcf
21 Apr, 2021 5 commits
- Add optional random stride to the video input pipeline. · e3774f8c
  Dan Kondratyuk authored Apr 21, 2021
```
PiperOrigin-RevId: 369747553
```
  e3774f8c
- Internal change · 1b1fca6d
  Yeqing Li authored Apr 21, 2021
```
PiperOrigin-RevId: 369741002
```
  1b1fca6d
- Adds a utility package for creating fake data for test. · 1446b8cc
  Yeqing Li authored Apr 21, 2021
```
PiperOrigin-RevId: 369712987
```
  1446b8cc
- Internal change · 9239c294
  A. Unique TensorFlower authored Apr 21, 2021
```
PiperOrigin-RevId: 369697787
```
  9239c294
- Update `Config` class doc to warn the auto conversion of dict to `Config` class. · 7678a1e9
  Hongkun Yu authored Apr 20, 2021
```
PiperOrigin-RevId: 369588823
```
  7678a1e9
20 Apr, 2021 2 commits
- Internal change · 6e65d027
  A. Unique TensorFlower authored Apr 20, 2021
```
PiperOrigin-RevId: 369459880
```
  6e65d027
- Expose learning rate schedule classes. · 3c46cfb5
  Hongkun Yu authored Apr 19, 2021
```
PiperOrigin-RevId: 369375299
```
  3c46cfb5
19 Apr, 2021 3 commits
- Add mobilebert to official/README.md · cf86fd97
  Chen Chen authored Apr 19, 2021
```
PiperOrigin-RevId: 369316688
```
  cf86fd97
- Internal change · c4c567ae
  Yeqing Li authored Apr 19, 2021
```
PiperOrigin-RevId: 369249071
```
  c4c567ae
- Add `include_example_field` into `SentencePredictionTextDataLoader` so that we... · c430556b
  A. Unique TensorFlower authored Apr 19, 2021
```
Add `include_example_field` into `SentencePredictionTextDataLoader` so that we can use the data loader in the predict step of `SentencePrediction` task.

PiperOrigin-RevId: 369215827
```
  c430556b
17 Apr, 2021 2 commits

Chen Chen authored Apr 16, 2021

tensorflow.python.framework.errors_impl.NotFoundError: /usr/local/lib/python3.6/dist-packages/tensorflow_text/python/metrics/_text_similarity_metric_ops.so: undefined symbol: _ZN10tensorflow6StatusC1ENS_5error4CodeEN4absl12lts_2021032411string_viewEOSt6vectorINS_10StackFrameESaIS7_EE

PiperOrigin-RevId: 368978629

ffaa4035

Internal change to image classification. · 8fa62b84
Fan Yang authored Apr 16, 2021
```
PiperOrigin-RevId: 368957441
```
8fa62b84

16 Apr, 2021 3 commits
- Internal change. · c2e19c97
  Fan Yang authored Apr 16, 2021
```
PiperOrigin-RevId: 368935233
```
  c2e19c97
- Internal change · 81a19f62
  A. Unique TensorFlower authored Apr 16, 2021
```
PiperOrigin-RevId: 368868424
```
  81a19f62
- Adds trainer and checkpoint exporter as the arguments of the run_experiment functions. · e3704ce2
  Yeqing Li authored Apr 15, 2021
```
PiperOrigin-RevId: 368778443
```
  e3704ce2
15 Apr, 2021 3 commits
- Internal change · 85a6db17
  Scott Zhu authored Apr 14, 2021
```
PiperOrigin-RevId: 368573031
```
  85a6db17
- Internal change · d0879611
  Scott Zhu authored Apr 14, 2021
```
PiperOrigin-RevId: 368571725
```
  d0879611
- Internal change · 147b46e9
  Yeqing Li authored Apr 14, 2021
```
PiperOrigin-RevId: 368535973
```
  147b46e9
14 Apr, 2021 2 commits
- Internal change · 2b68aa95
  Hongkun Yu authored Apr 14, 2021
```
PiperOrigin-RevId: 368535232
```
  2b68aa95
- Change activation/add to a keras layer form for the ResNet50, MobilenetV2 default case. · 76476cd9
  Jaehong Kim authored Apr 14, 2021
```
This CL changes ResidualBlock and InvertedBottleneckBlock.

PiperOrigin-RevId: 368383954
```
  76476cd9
13 Apr, 2021 7 commits

Internal change · b1aa44d9
A. Unique TensorFlower authored Apr 13, 2021
```
PiperOrigin-RevId: 368260712
```
b1aa44d9
Fix parse_configuration using "in" with ParseConfigOptions · 782a0299
Hongkun Yu authored Apr 13, 2021
```
PiperOrigin-RevId: 368257425
```
782a0299
Internal change · 87796817
Rebecca Chen authored Apr 12, 2021
```
PiperOrigin-RevId: 368157180
```
87796817
Internal change · 0c814fa2
Jaehong Kim authored Apr 12, 2021
```
PiperOrigin-RevId: 368130039
```
0c814fa2
Add more pytype checking. · f8f4845c
Fan Yang authored Apr 12, 2021
```
PiperOrigin-RevId: 368129317
```
f8f4845c

Use nonexperimental mixed precision API. · 4334a892

Reed Wanderman-Milne authored Apr 12, 2021

This replaces symbols in tf.keras.mixed_precision.experimental with the corresponding nonexperimental symbols. In some cases, passing a Policy is replaced with passing a policy name for conciseness.

Additionally, for the Shakespeare model, the loss_scale flag is removed, since supporting it with the nonexperimental API is slightly more verbose and it is recommended users use the default loss scale.

PiperOrigin-RevId: 368123944

4334a892

Internal change · dbf629c4
Hongkun Yu authored Apr 12, 2021
```
PiperOrigin-RevId: 368122127
```
dbf629c4

12 Apr, 2021 5 commits

Use nonexperimental mixed precision API for official models. · 0d8f9807

Reed Wanderman-Milne authored Apr 12, 2021

For all modified calls to set_mixed_precision_policy(), the loss_scale argument was removed, as it cannot be passed if the nonexperimental API is used. For all such callers, the loss_scale is later used to explicitly create a LossScaleOptimizer, so removing the argument has no impact.

Switching to the non-experimental LossScaleOptimizer has no effect, as it has near identical behavior and all isinstance checks within the official models check for the non-experimental version.

PiperOrigin-RevId: 368101975

0d8f9807

Internal change · 0338ca4b
Tianjian Meng authored Apr 12, 2021
```
PiperOrigin-RevId: 368070382
```
0338ca4b
Internal change · 1bbd359d
Reed Wanderman-Milne authored Apr 12, 2021
```
PiperOrigin-RevId: 368067415
```
1bbd359d
Internal change · ecbc3fd8
A. Unique TensorFlower authored Apr 12, 2021
```
PiperOrigin-RevId: 368040234
```
ecbc3fd8
Internal change · 6f917c69
Abdullah Rashwan authored Apr 12, 2021
```
PiperOrigin-RevId: 368036370
```
6f917c69

09 Apr, 2021 1 commit

Remove dynamic_loss_scale argument to define_performance. · e353e4e5

Reed Wanderman-Milne authored Apr 09, 2021

All models which support loss scaling support dynamic loss scaling, so the argument has no purpose. It used to be that some models scaled the loss manually instead of using a LossScaleOptimizer, and so did not support dynamic loss scaling.

PiperOrigin-RevId: 367719521

e353e4e5