提交 · f079ed2ec359bc0f011bf5d48fc77a6900f360e2 · 曾经的那一瞬间 / Models

06 12月, 2019 2 次提交
- H
  Internal change · f079ed2e
  由 Hongkun Yu 提交于 12月 05, 2019
```
PiperOrigin-RevId: 284016229
```
  f079ed2e
- Y
  Imports the mask-rcnn config. · 59393a8d
  由 Yeqing Li 提交于 12月 05, 2019
```
PiperOrigin-RevId: 283987800
```
  59393a8d
05 12月, 2019 8 次提交
- J
  Internal change · c115444f
  由 Jaehong Kim 提交于 12月 05, 2019
```
PiperOrigin-RevId: 283962490
```
  c115444f
- H
  Remove tweaked benchmark for no_dist_strat in function mode. · ef8aed79
  由 Haoyu Zhang 提交于 12月 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
- C
  Make albert_transformer_encoder*.py public available · ccf64277
  由 Chen Chen 提交于 12月 04, 2019
```
PiperOrigin-RevId: 283885197
```
  ccf64277
- A
  Moved set_gpu_thread_mode_and_count from vision/image_classification/common.py to · f926be0a
  由 A. Unique TensorFlower 提交于 12月 04, 2019
```
utils/misc/keras_utils

PiperOrigin-RevId: 283885102
```
  f926be0a
- Y
  Code cleanup. · 91a1ce9b
  由 Yeqing Li 提交于 12月 04, 2019
```
PiperOrigin-RevId: 283837279
```
  91a1ce9b
- P
  Internal change · 5b25005c
  由 Pengchong Jin 提交于 12月 04, 2019
```
PiperOrigin-RevId: 283817737
```
  5b25005c
- A
  Internal change · f3a61a49
  由 A. Unique TensorFlower 提交于 12月 04, 2019
```
PiperOrigin-RevId: 283778598
```
  f3a61a49
- Y
  Internal change · 4270e416
  由 Yeqing Li 提交于 12月 04, 2019
```
PiperOrigin-RevId: 283770513
```
  4270e416
04 12月, 2019 3 次提交
- P
  Fix MobileNet example notebook issue (#4543) · ad56514f
  由 Philip Yang 提交于 12月 03, 2019
```
Fix variable name issue in MobileNet example notebook.
Make sure all `base_name` variables are renamed to `checkpoint_name`.
```
  ad56514f
- A
  Internal change · 6a50c338
  由 A. Unique TensorFlower 提交于 12月 03, 2019
```
PiperOrigin-RevId: 283648734
```
  6a50c338
- H
  Internal change · 1d3cd3cf
  由 Hongkun Yu 提交于 12月 03, 2019
```
PiperOrigin-RevId: 283622345
```
  1d3cd3cf
03 12月, 2019 4 次提交
- A
  Internal change · b478430d
  由 A. Unique TensorFlower 提交于 12月 02, 2019
```
PiperOrigin-RevId: 283449673
```
  b478430d
- H
  Fix bug introduced. · 281d1757
  由 Hongkun Yu 提交于 12月 02, 2019
```
PiperOrigin-RevId: 283447458
```
  281d1757
- A
  Internal change · 1dffde32
  由 A. Unique TensorFlower 提交于 12月 02, 2019
```
PiperOrigin-RevId: 283377414
```
  1dffde32
- C
  Move tf2_encoder_checkpoint_converter to public. · da228b42
  由 Chen Chen 提交于 12月 02, 2019
```
PiperOrigin-RevId: 283374562
```
  da228b42
02 12月, 2019 1 次提交
- H
  Internal change · 494cf0b3
  由 Hongkun Yu 提交于 12月 01, 2019
```
PiperOrigin-RevId: 283266705
```
  494cf0b3
30 11月, 2019 2 次提交

A
Merge pull request #7884 from miraliahmadli:master · 676c23e6
由 A. Unique TensorFlower 提交于 11月 29, 2019
```
PiperOrigin-RevId: 283114824
```
676c23e6

Update vid2depth/README.md: use raw.githubcontent (#7881) · d49cb9ef

由 ReGuess 提交于 11月 30, 2019

The wget command for the KITTI dataset in the currently downloads the html for GitHub's fancy display of the text file. This change fixes that so it instead downloads the raw txt file (no more trying to resolve host '<')

d49cb9ef

29 11月, 2019 1 次提交
- M
  
  Fixed tf.compat.v1 bug · a5c96853
  由 Mirali Ahmadli 提交于 11月 29, 2019
  
  a5c96853
28 11月, 2019 7 次提交
- S
  Internal change · 0effd158
  由 Sai Ganesh Bandiatmakuri 提交于 11月 27, 2019
```
PiperOrigin-RevId: 282857541
```
  0effd158
- P
  Clean up the mask utils. · 61d6173d
  由 Pengchong Jin 提交于 11月 27, 2019
```
PiperOrigin-RevId: 282825837
```
  61d6173d
- H
  Enabled performance related parameters for Transformer: all_reduce_alg, · bd73fdfe
  由 Hongkun Yu 提交于 11月 27, 2019
```
enable_eager, tf_gpu_thread_mode and datasets_num_private_threads

PiperOrigin-RevId: 282804724
```
  bd73fdfe
- S
  Inject enable_runtime_flags into benchmarks and add train_steps to shakespeare_main. · 808b2f5e
  由 Sai Ganesh Bandiatmakuri 提交于 11月 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
- H
  Update BERT readme with new checkpoints fully ready. · 09a639bf
  由 Hongkun Yu 提交于 11月 27, 2019
```
PiperOrigin-RevId: 282798626
```
  09a639bf
- A
  Enabled performance related parameters for Transformer: all_reduce_alg, · 18df1a09
  由 A. Unique TensorFlower 提交于 11月 27, 2019
```
enable_eager, tf_gpu_thread_mode and datasets_num_private_threads

PiperOrigin-RevId: 282797514
```
  18df1a09
- P
  Internal change · ae989abc
  由 Pengchong Jin 提交于 11月 27, 2019
```
PiperOrigin-RevId: 282795425
```
  ae989abc
27 11月, 2019 3 次提交
- J
  Add tensorflow-model-optimization dependency. · ad3ff44b
  由 Jaehong Kim 提交于 11月 26, 2019
```
PiperOrigin-RevId: 282695365
```
  ad3ff44b
- H
  Remove 'default' in get_distribution_strategy which is complex and error-prone · 04256053
  由 Hongkun Yu 提交于 11月 26, 2019
```
PiperOrigin-RevId: 282669615
```
  04256053
- A
  Internal change · 30165f86
  由 A. Unique TensorFlower 提交于 11月 26, 2019
```
PiperOrigin-RevId: 282650416
```
  30165f86
26 11月, 2019 3 次提交
- H
  Implement CachedAttention layer. This is useful for decoders. · 0b579232
  由 Hongkun Yu 提交于 11月 25, 2019
```
PiperOrigin-RevId: 282439719
```
  0b579232
- S
  Inject enable_runtime_flags into benchmarks. · bcce419a
  由 Sai Ganesh Bandiatmakuri 提交于 11月 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
- C
  Completely remove "use_keras_bert_for_squad" FLAG. · 9c50e961
  由 Chen Chen 提交于 11月 25, 2019
```
PiperOrigin-RevId: 282384787
```
  9c50e961
25 11月, 2019 2 次提交
- Z
  Internal change · 5b2575c2
  由 Zongwei Zhou 提交于 11月 24, 2019
```
PiperOrigin-RevId: 282293668
```
  5b2575c2
- B
  
  minor sp: MobilenNet --> MobileNet (#7768) · 1a40ebdd
  由 brett koonce 提交于 11月 24, 2019
  
  1a40ebdd
23 11月, 2019 4 次提交
- H
  Bring back gather_indexes · ff8ee1ae
  由 Hongkun Yu 提交于 11月 22, 2019
```
PiperOrigin-RevId: 282103499
```
  ff8ee1ae
- C
  Internal change · 83f0a576
  由 Chen Chen 提交于 11月 22, 2019
```
PiperOrigin-RevId: 282096004
```
  83f0a576
- Y
  Internal change · 986ffac4
  由 Yeqing Li 提交于 11月 22, 2019
```
PiperOrigin-RevId: 282065024
```
  986ffac4
- Y
  
  Update readme and add mobilenet v3 reference (#7847) · a9387332
  由 Yukun Zhu 提交于 11月 22, 2019
  
  a9387332

曾经的那一瞬间 / Models 大约 1 年 前同步成功

曾经的那一瞬间 / Models
大约 1 年前同步成功