Commits · 1b1fca6d7fb2908fe0cf26025e92a312af63c904 · ModelZoo / ResNet50_tensorflow

"vscode:/vscode.git/clone" did not exist on "2cd7e181dddd825862e8007b3709e12339e62f03"

21 Apr, 2021 4 commits
- 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 5 commits
- Correct 404 link for object detection tutorial (#9210) · acbd8837
  Andrea Garritano authored Apr 20, 2021
  
  acbd8837
- [object detection] Bug fixes in colab tutorial (#9725) · 8f70ff1c
  Tyler Ganter authored Apr 20, 2021
```
* OD tutorial colab working

* fixing install instructions link
```
  8f70ff1c
- Internal change · 6e65d027
  A. Unique TensorFlower authored Apr 20, 2021
```
PiperOrigin-RevId: 369459880
```
  6e65d027
- Surgical update to CenterNet keypoint postprocessing to use bounding boxes if... · 94618de6
  Ronny Votel authored Apr 20, 2021
```
Surgical update to CenterNet keypoint postprocessing to use bounding boxes if present. This is shown to reduce keypoint snapping across individuals.

PiperOrigin-RevId: 369432018
```
  94618de6
- Expose learning rate schedule classes. · 3c46cfb5
  Hongkun Yu authored Apr 19, 2021
```
PiperOrigin-RevId: 369375299
```
  3c46cfb5
19 Apr, 2021 4 commits
- Add mobilebert to official/README.md · cf86fd97
  Chen Chen authored Apr 19, 2021
```
PiperOrigin-RevId: 369316688
```
  cf86fd97
- Update CODEOWNERS · 067817b9
  Jaeyoun Kim authored Apr 19, 2021
```
Add owners for new projects (assemblenet, deepmac_maskrcnn, simclr)
```
  067817b9
- 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 5 commits

Internal change. · c2e19c97
Fan Yang authored Apr 16, 2021
```
PiperOrigin-RevId: 368935233
```
c2e19c97
Add efficientnet to community (#9894) · 127c9d80
Pablo Ribalta Lorenzo authored Apr 16, 2021
```
Signed-off-by: Pablo Ribalta Lorenzo <pribalta@nvidia.com>
```
127c9d80

Global model supporting functions (#9901) · 79946a70

Jekaterina Jaroslavceva authored Apr 16, 2021

* Dataset utilities added.

* Global model definition

* Dataset modules added.

* Dataset modules fix.

* global features model training added

* global features fix

* Test dataset update

* PR fixes

* repo sync

* repo sync

* Syncing 2

* Syncing 2

* Added global model supporting modules

* code style fixes

* Minor style fixes

79946a70

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 5 commits

Fixed deprecated tf.flags (#9912) · a56d7ab3
Leandro Schelb authored Apr 14, 2021

a56d7ab3
Internal change · 2b68aa95
Hongkun Yu authored Apr 14, 2021
```
PiperOrigin-RevId: 368535232
```
2b68aa95

Call set_keypoint_visibilities on the non-expanded versions of the detection... · 8a16208b

A. Unique TensorFlower authored Apr 14, 2021

Call set_keypoint_visibilities on the non-expanded versions of the detection and groundtruth keypoints. set_keypoint_visibilities expects a rank-3 tensor, and was being provided a rank-4 tensor. This had the unintended effect of creating a keypoint visibilities tensor of the wrong shape, resulting in only 2 keypoints being visualized.

PiperOrigin-RevId: 368458273

8a16208b

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

Fix the typo-ed name "remove_unecessary_ema" -> "remove_unnecessary_ema" and... · bf2d1354

Liangzhe Yuan authored Apr 13, 2021

Fix the typo-ed name "remove_unecessary_ema" -> "remove_unnecessary_ema" and also re-implement it with deep dictionary copy.

PiperOrigin-RevId: 368338501

bf2d1354

13 Apr, 2021 11 commits

File movings (#9899) · 3dacd474

Jekaterina Jaroslavceva authored Apr 13, 2021

* Dataset utilities added.

* Global model definition

* Dataset modules added.

* Dataset modules fix.

* global features model training added

* global features fix

* Test dataset update

* PR fixes

* repo sync

* repo sync

* modules moving

* Removed unnecessary modules

* Removed unnecessary files

3dacd474

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
Replacing tf.gather_nd since reshaping and applying tf.gather executes faster during training. · 3d9ae6de
Ronny Votel authored Apr 13, 2021
```
PiperOrigin-RevId: 368217758
```
3d9ae6de
Clip values in strided box predictions. · e5459a6b
Vighnesh Birodkar authored Apr 13, 2021
```
PiperOrigin-RevId: 368215027
```
e5459a6b
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

Use nonexperimental mixed precision API. · 19d18c00

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.

PiperOrigin-RevId: 368123944

19d18c00

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

12 Apr, 2021 1 commit

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