Commits · 26bbda7383887b6191b74b26942b0a2fade7d8c1 · ModelZoo / ResNet50_tensorflow

11 Dec, 2019 1 commit
- Internal change · 26bbda73
  David Chen authored Dec 10, 2019
```
PiperOrigin-RevId: 284874717
```
  26bbda73
10 Dec, 2019 3 commits
- A more accurate implementation of mask pasting. · 38e48f91
  Pengchong Jin authored Dec 10, 2019
```
PiperOrigin-RevId: 284854952
```
  38e48f91
- Internal change · 68104ce3
  A. Unique TensorFlower authored Dec 10, 2019
```
PiperOrigin-RevId: 284805626
```
  68104ce3
- Internal change · f7fd59b8
  Chen Chen authored Dec 10, 2019
```
PiperOrigin-RevId: 284792715
```
  f7fd59b8
09 Dec, 2019 1 commit
- Add sentence piece tokenizer in tokenization.py · 558bab5d
  Chen Chen authored Dec 09, 2019
```
PiperOrigin-RevId: 284624714
```
  558bab5d
06 Dec, 2019 5 commits
- Internal Change · 440f31ab
  Chen Chen authored Dec 06, 2019
```
PiperOrigin-RevId: 284278191
```
  440f31ab
- Internal cleanup. · 8a8baded
  Yeqing Li authored Dec 06, 2019
```
PiperOrigin-RevId: 284259997
```
  8a8baded
- Internal change · 6e847fbd
  Chen Chen authored Dec 06, 2019
```
PiperOrigin-RevId: 284203261
```
  6e847fbd
- Remove code that disables pin-to-host optimization. It's disabled by default... · 6f64dcb5
  Haoyu Zhang authored Dec 05, 2019
```
Remove code that disables pin-to-host optimization. It's disabled by default and we already have XLA presubmit that will catch any issues when the optimization is reenabled.

PiperOrigin-RevId: 284078252
```
  6f64dcb5
- Using better version of l2 loss to avoid reshape, concat and split ops. · c0bdb378
  A. Unique TensorFlower authored Dec 05, 2019
```
Also adding support for CTL mode in the borg file.

PiperOrigin-RevId: 284075404
```
  c0bdb378
05 Dec, 2019 6 commits
- Internal change · f079ed2e
  Hongkun Yu authored Dec 05, 2019
```
PiperOrigin-RevId: 284016229
```
  f079ed2e
- Imports the mask-rcnn config. · 59393a8d
  Yeqing Li authored Dec 05, 2019
```
PiperOrigin-RevId: 283987800
```
  59393a8d
- Internal change · c115444f
  Jaehong Kim authored Dec 05, 2019
```
PiperOrigin-RevId: 283962490
```
  c115444f
- Remove tweaked benchmark for no_dist_strat in function mode. · ef8aed79
  Haoyu Zhang authored Dec 04, 2019
```
No need to test explicit_gpu_placement + no_dist_strat + function case, since the performance issue is already fixed by unifying the Keras codepaths. However, there is still performance gap using the explicit_gpu_placement flag when run_eagerly.

PiperOrigin-RevId: 283890008
```
  ef8aed79
- Make albert_transformer_encoder*.py public available · ccf64277
  Chen Chen authored Dec 04, 2019
```
PiperOrigin-RevId: 283885197
```
  ccf64277
- Moved set_gpu_thread_mode_and_count from vision/image_classification/common.py to · f926be0a
  A. Unique TensorFlower authored Dec 04, 2019
```
utils/misc/keras_utils

PiperOrigin-RevId: 283885102
```
  f926be0a
04 Dec, 2019 5 commits
- Code cleanup. · 91a1ce9b
  Yeqing Li authored Dec 04, 2019
```
PiperOrigin-RevId: 283837279
```
  91a1ce9b
- Internal change · 5b25005c
  Pengchong Jin authored Dec 04, 2019
```
PiperOrigin-RevId: 283817737
```
  5b25005c
- Internal change · f3a61a49
  A. Unique TensorFlower authored Dec 04, 2019
```
PiperOrigin-RevId: 283778598
```
  f3a61a49
- Internal change · 4270e416
  Yeqing Li authored Dec 04, 2019
```
PiperOrigin-RevId: 283770513
```
  4270e416
- Internal change · 6a50c338
  A. Unique TensorFlower authored Dec 03, 2019
```
PiperOrigin-RevId: 283648734
```
  6a50c338
03 Dec, 2019 3 commits
- Internal change · 1d3cd3cf
  Hongkun Yu authored Dec 03, 2019
```
PiperOrigin-RevId: 283622345
```
  1d3cd3cf
- Internal change · b478430d
  A. Unique TensorFlower authored Dec 02, 2019
```
PiperOrigin-RevId: 283449673
```
  b478430d
- Fix bug introduced. · 281d1757
  Hongkun Yu authored Dec 02, 2019
```
PiperOrigin-RevId: 283447458
```
  281d1757
02 Dec, 2019 3 commits
- Internal change · 1dffde32
  A. Unique TensorFlower authored Dec 02, 2019
```
PiperOrigin-RevId: 283377414
```
  1dffde32
- Move tf2_encoder_checkpoint_converter to public. · da228b42
  Chen Chen authored Dec 02, 2019
```
PiperOrigin-RevId: 283374562
```
  da228b42
- Internal change · 494cf0b3
  Hongkun Yu authored Dec 01, 2019
```
PiperOrigin-RevId: 283266705
```
  494cf0b3
29 Nov, 2019 1 commit
- Fixed tf.compat.v1 bug · a5c96853
  Mirali Ahmadli authored Nov 29, 2019
  
  a5c96853
28 Nov, 2019 1 commit
- Internal change · 0effd158
  Sai Ganesh Bandiatmakuri authored Nov 27, 2019
```
PiperOrigin-RevId: 282857541
```
  0effd158
27 Nov, 2019 8 commits
- Clean up the mask utils. · 61d6173d
  Pengchong Jin authored Nov 27, 2019
```
PiperOrigin-RevId: 282825837
```
  61d6173d
- Enabled performance related parameters for Transformer: all_reduce_alg, · bd73fdfe
  Hongkun Yu authored Nov 27, 2019
```
enable_eager, tf_gpu_thread_mode and datasets_num_private_threads

PiperOrigin-RevId: 282804724
```
  bd73fdfe
- Inject enable_runtime_flags into benchmarks and add train_steps to shakespeare_main. · 808b2f5e
  Sai Ganesh Bandiatmakuri authored Nov 27, 2019
```
This will help general debugging by enabling custom execution with
--benchmark_method_flags.

E.g --benchmark_method_flags=train_steps=7 will run the benchmark for only 7
steps without modifying benchmark code.

PiperOrigin-RevId: 282799341
```
  808b2f5e
- Update BERT readme with new checkpoints fully ready. · 09a639bf
  Hongkun Yu authored Nov 27, 2019
```
PiperOrigin-RevId: 282798626
```
  09a639bf
- Enabled performance related parameters for Transformer: all_reduce_alg, · 18df1a09
  A. Unique TensorFlower authored Nov 27, 2019
```
enable_eager, tf_gpu_thread_mode and datasets_num_private_threads

PiperOrigin-RevId: 282797514
```
  18df1a09
- Internal change · ae989abc
  Pengchong Jin authored Nov 27, 2019
```
PiperOrigin-RevId: 282795425
```
  ae989abc
- Add tensorflow-model-optimization dependency. · ad3ff44b
  Jaehong Kim authored Nov 26, 2019
```
PiperOrigin-RevId: 282695365
```
  ad3ff44b
- Remove 'default' in get_distribution_strategy which is complex and error-prone · 04256053
  Hongkun Yu authored Nov 26, 2019
```
PiperOrigin-RevId: 282669615
```
  04256053
26 Nov, 2019 1 commit
- Internal change · 30165f86
  A. Unique TensorFlower authored Nov 26, 2019
```
PiperOrigin-RevId: 282650416
```
  30165f86
25 Nov, 2019 2 commits

Implement CachedAttention layer. This is useful for decoders. · 0b579232
Hongkun Yu authored Nov 25, 2019
```
PiperOrigin-RevId: 282439719
```
0b579232

Inject enable_runtime_flags into benchmarks. · bcce419a

Sai Ganesh Bandiatmakuri authored Nov 25, 2019

This will help general debugging by enabling custom execution with --benchmark_method_steps.

E.g --benchmark_method_steps=train_steps=7 will run the benchmark for only 7 steps without modifying benchmark code.

PiperOrigin-RevId: 282396875

bcce419a