Commits · 23837ecfe58361e84106956cbc445ecc89b035c5 · ModelZoo / ResNet50_tensorflow

13 Jul, 2021 4 commits

Jekaterina Jaroslavceva authored Jul 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

* Syncing 2

* Syncing 2

* Global model definition added

* Global model definition added, synced

* Adding global model dataset related modules

* Global model training

* tensorboard module added

* linting issues fixed

* linting fixes.

* linting fixes.

* Fix for previous PR

* PR fixes

* Minor fixes

* Minor fixes

* Dataset download fix

* Comments fix

* sfm120k fix

* sfm120k fix

* names fix

* Update

* Update

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* PR fixes global model

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

23837ecf

Internal change · 5bac2ced
Rino Lee authored Jul 13, 2021
```
PiperOrigin-RevId: 384478988
```
5bac2ced
Internal change · 0e096a84
A. Unique TensorFlower authored Jul 12, 2021
```
PiperOrigin-RevId: 384358378
```
0e096a84
[Docs] Update supported TF-Vision models to the official/README.md · f2f4f2dc
Hongkun Yu authored Jul 12, 2021
```
PiperOrigin-RevId: 384356677
```
f2f4f2dc

12 Jul, 2021 3 commits
- Add control to turn-on/off BN for segmentation head 3D. · 19e1187a
  Fan Yang authored Jul 12, 2021
```
PiperOrigin-RevId: 384262412
```
  19e1187a
- Fix the issue when both numerator and denominator are 0, the dice score is... · b7e42e53
  Fan Yang authored Jul 12, 2021
```
Fix the issue when both numerator and denominator are 0, the dice score is incorrectly computed as 1, which should be 0.

PiperOrigin-RevId: 384257731
```
  b7e42e53
- Internal change. · 9cff6e3b
  Fan Yang authored Jul 12, 2021
```
PiperOrigin-RevId: 384257276
```
  9cff6e3b
10 Jul, 2021 2 commits
- Internal change · 0e74158f
  A. Unique TensorFlower authored Jul 10, 2021
```
PiperOrigin-RevId: 384018258
```
  0e74158f
- [efficient] Fix is_short_seq order so that we can also apply feature transform... · 5f23689e
  Frederick Liu authored Jul 09, 2021
```
[efficient] Fix is_short_seq order so that we can also apply feature transform and apply softmax afterwards.

PiperOrigin-RevId: 383967806
```
  5f23689e
09 Jul, 2021 4 commits

Refactor decoder factory to allow registering other decoders. · dbaec326
Fan Yang authored Jul 09, 2021
```
PiperOrigin-RevId: 383944185
```
dbaec326
Internal change · bc71d8e9
Dan Kondratyuk authored Jul 09, 2021
```
PiperOrigin-RevId: 383861528
```
bc71d8e9
Internal change · cb9aeaf4
Dan Kondratyuk authored Jul 09, 2021
```
PiperOrigin-RevId: 383856955
```
cb9aeaf4

With float16, always use LossScaleOptimizer. · 21286f77

Reed Wanderman-Milne authored Jul 08, 2021

Before, it was too easy to accidentally forget to set runtime.loss_scale, which had to always be done if mixed precision is used, otherwise the model would converge to worse accuracy. Now, all that needs to be done to use mixed precision is to set runtime.mixed_precision_dtype=float16.

PiperOrigin-RevId: 383767033

21286f77

08 Jul, 2021 5 commits
- Mask boxes with centers outside of the true_image_shape. · c7d4dd39
  A. Unique TensorFlower authored Jul 08, 2021
```
PiperOrigin-RevId: 383699499
```
  c7d4dd39
- Internal change · 07589c47
  A. Unique TensorFlower authored Jul 08, 2021
```
PiperOrigin-RevId: 383679175
```
  07589c47
- Internal change · 6c835dd6
  Hongkun Yu authored Jul 08, 2021
```
PiperOrigin-RevId: 383667471
```
  6c835dd6
- Fix typo in TransformerEncoderBlock docstring · 420e8179
  Philip Pham authored Jul 08, 2021
```
PiperOrigin-RevId: 383660517
```
  420e8179
- Add explanation for anchor box configuration · e71d67c3
  Scott Main authored Jul 08, 2021
```
PiperOrigin-RevId: 383651497
```
  e71d67c3
07 Jul, 2021 4 commits

Internal change · 178fdeb2
Abdullah Rashwan authored Jul 07, 2021
```
PiperOrigin-RevId: 383505893
```
178fdeb2
Internal change · eedfc5f6
Hongkun Yu authored Jul 07, 2021
```
PiperOrigin-RevId: 383505455
```
eedfc5f6
Standardize CenterNet mobilenet v2 FPN with new checkpoint API. · 2e9c6046
Vighnesh Birodkar authored Jul 07, 2021
```
PiperOrigin-RevId: 383469336
```
2e9c6046

Dataset modules (#9903) · fa038f86

Jekaterina Jaroslavceva authored Jul 07, 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

* Global model definition added

* Global model definition added, synced

* Adding global model dataset related modules

* Global model training

* tensorboard module added

* linting issues fixed

* linting fixes.

* linting fixes.

* Fix for previous PR

* PR fixes

* Minor fixes

* Minor fixes

* Dataset download fix

* Comments fix

* sfm120k fix

* sfm120k fix

* names fix

* Update

* Update

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* PR fixes

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

* Merge branch 'global_model_training'

# Conflicts:
#	research/delf/delf/python/datasets/generic_dataset.py
#	research/delf/delf/python/datasets/generic_dataset_test.py
#	research/delf/delf/python/datasets/sfm120k/__init__.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k.py
#	research/delf/delf/python/datasets/sfm120k/sfm120k_test.py
#	research/delf/delf/python/datasets/tuples_dataset.py
#	research/delf/delf/python/datasets/tuples_dataset_test.py
#	research/delf/delf/python/training/global_features/__init__.py
#	research/delf/delf/python/training/global_features/train.py
#	research/delf/delf/python/training/model/global_model.py
#	research/delf/delf/python/training/model/global_model_test.py
#	research/delf/delf/python/training/tensorboard_utils.py

fa038f86

05 Jul, 2021 1 commit
- [core] Remove adafactor factory test since it's not implemented which causes external test to fail. · 2924510a
  Frederick Liu authored Jul 04, 2021
```
PiperOrigin-RevId: 383066915
```
  2924510a
04 Jul, 2021 1 commit
- [nlp] Support norm_first option for Bert/Bigbird/Kernel encoders. · 72c47570
  Frederick Liu authored Jul 03, 2021
```
PiperOrigin-RevId: 382960239
```
  72c47570
03 Jul, 2021 1 commit
- Internal change · b3483b39
  Frederick Liu authored Jul 02, 2021
```
PiperOrigin-RevId: 382846192
```
  b3483b39
02 Jul, 2021 3 commits
- Remove private Keras API usage. · affbaa60
  Dan Kondratyuk authored Jul 02, 2021
```
PiperOrigin-RevId: 382833629
```
  affbaa60
- Minor change to mask target assigner to optionally scale boxes before assigning dense loss weights. · 170c71cc
  Ronny Votel authored Jul 02, 2021
```
PiperOrigin-RevId: 382773317
```
  170c71cc
- Internal change · 13100073
  A. Unique TensorFlower authored Jul 01, 2021
```
PiperOrigin-RevId: 382655843
```
  13100073
01 Jul, 2021 11 commits
- Change activation/add/upsampling/identity to a keras layer form for the spinenet_mobile. · 00024735
  Jaehong Kim authored Jul 01, 2021
```
Added use_keras_upsampling_2d flags to use keras upsampling layer instead of optimized custom spatial_transform_ops.

PiperOrigin-RevId: 382626602
```
  00024735
- Revert causal padding until new checkpoints are trained. · cbe77ec0
  Dan Kondratyuk authored Jul 01, 2021
```
PiperOrigin-RevId: 382606601
```
  cbe77ec0
- Internal change · c3b4fa95
  A. Unique TensorFlower authored Jul 01, 2021
```
PiperOrigin-RevId: 382571340
```
  c3b4fa95
- Update README.md · c3429389
  Jaeyoun Kim authored Jul 01, 2021
```
Update the section for citation
```
  c3429389
- Internal change · ec664434
  Abdullah Rashwan authored Jul 01, 2021
```
PiperOrigin-RevId: 382561014
```
  ec664434
- Update README.md · 7bb84024
  Jaeyoun Kim authored Jul 01, 2021
```
Revise the citation section
```
  7bb84024
- Update README.md · 5744dd30
  Jaeyoun Kim authored Jul 01, 2021
  
  5744dd30
- Update README.md · a158dce0
  Jaeyoun Kim authored Jul 01, 2021
```
Add information how to cite TensorFlow Model Garden
```
  a158dce0
- Merge pull request #10081 from Atze00:padding_causal_convolution · 9b012b52
  A. Unique TensorFlower authored Jul 01, 2021
```
PiperOrigin-RevId: 382547358
```
  9b012b52
- Internal change · 4f655acd
  Abdullah Rashwan authored Jul 01, 2021
```
PiperOrigin-RevId: 382462213
```
  4f655acd
- Internal change · b1d973ea
  Abdullah Rashwan authored Jun 30, 2021
```
PiperOrigin-RevId: 382422745
```
  b1d973ea
30 Jun, 2021 1 commit
- Internal change · e515beb1
  Xianzhi Du authored Jun 30, 2021
```
PiperOrigin-RevId: 382408293
```
  e515beb1