Commits · 9485aa1dcf2c2e1ebaede0af0283796dcfb01911 · ModelZoo / ResNet50_tensorflow

13 Apr, 2021 1 commit

Use nonexperimental mixed precision API. · c0ac8d1c

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

c0ac8d1c

23 Mar, 2021 1 commit
- Parallelize the batching of the Shakespeare benchmark. · 0ffd6ab0
  Jay Shi authored Mar 23, 2021
```
PiperOrigin-RevId: 364541922
```
  0ffd6ab0
13 Sep, 2020 1 commit
- [Clean up] Consolidate distribution utils. · faf7a4f5
  Hongkun Yu authored Sep 12, 2020
```
PiperOrigin-RevId: 331359058
```
  faf7a4f5
10 Aug, 2020 1 commit
- Remove benchmark folder from the master branch. They are stale. (#9085) · c5e107ff
  Hongkun Yu authored Aug 10, 2020
  
  c5e107ff
19 May, 2020 1 commit
- [Clean up] Remove enable_eager in the session config: Model garden is TF2 only now. · c2666cea
  Hongkun Yu authored May 19, 2020
```
Remove is_v2_0

PiperOrigin-RevId: 312336907
```
  c2666cea
17 Apr, 2020 1 commit
- Remove hooks. · 2ddd716e
  Hongkun Yu authored Apr 16, 2020
```
PiperOrigin-RevId: 306994199
```
  2ddd716e
04 Apr, 2020 1 commit
- Move shakespeare to benchmark folder. · 17eee4c4
  Hongkun Yu authored Apr 03, 2020
```
PiperOrigin-RevId: 304729158
```
  17eee4c4
05 Mar, 2020 1 commit
- Remove force_v2_in_keras_compile. experimental_run_tf_function is no-op now. · d3d7f15f
  Hongkun Yu authored Mar 05, 2020
```
PiperOrigin-RevId: 299160422
```
  d3d7f15f
27 Nov, 2019 1 commit

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

06 Nov, 2019 1 commit
- Internal change · 02cc984e
  A. Unique TensorFlower authored Nov 05, 2019
```
PiperOrigin-RevId: 278754959
```
  02cc984e
08 Oct, 2019 1 commit
- Port shakespeare benchmark to TF2 · d0f21f25
  George Karpenkov authored Oct 08, 2019
```
PiperOrigin-RevId: 273527676
```
  d0f21f25
30 Sep, 2019 1 commit
- Remove dead code from shakespeare_main model · 6d6ab9ca
  George Karpenkov authored Sep 30, 2019
```
PiperOrigin-RevId: 272077584
```
  6d6ab9ca
09 Sep, 2019 1 commit

Unexpose some flags from models which do not use them. · e91c41c2

Reed Wanderman-Milne authored Sep 09, 2019

--stop_threshold, --num_gpu, --hooks, --export_dir, and --distribution_strategy have been unexposed from models which do not use them

PiperOrigin-RevId: 268032080

e91c41c2

04 Sep, 2019 1 commit
- Internal change · 154e8c46
  David Chen authored Sep 03, 2019
```
PiperOrigin-RevId: 267063328
```
  154e8c46
06 Aug, 2019 1 commit

Add non-CuDNN LSTM option and PerfZero benchmarks. (#7391) · 4d93d894

Taylor Robie authored Aug 06, 2019

* add non-CuDNN LSTM

* condition eager on v2 for no cudnn benchmark

* remove implementation specification

* add comments and adjust benchmarks

* re-add implementation=2

* update docstrings

4d93d894

31 Jul, 2019 1 commit
- Change to experimental_run_tf_function. (#7344) · a552e76a
  Toby Boyd authored Jul 31, 2019
  
  a552e76a
23 Jul, 2019 2 commits

Single execution path tests for ResNet50, ResNet56, NCF, and Shakespeare LSTM. (#7276) · 9d8c9aa4

Toby Boyd authored Jul 23, 2019

* Add force_run_distributed tests.

* Added enable_eager

* r/force_run_distributed/force_v2_in_keras_compile

* Adding force_v2 tests and FLAGs.

* Rename method to avoid conflict.

* Add cpu force_v2 tests.

* fix lint, wrap line.

* change to force_v2_in_keras_compile

* Update method name.

* Lower mlperf target to 0.736.

9d8c9aa4

add log_steps with faster logging for 8xGPU. (#7274) · 8390b362
Toby Boyd authored Jul 23, 2019

8390b362

18 Jul, 2019 1 commit

Refactor and add benchmarks as well as accuracy tests for GPU and CPU (#7248) · e0a2b8c3

Toby Boyd authored Jul 18, 2019

* Added benchmarks and common flags.

* Add cpu tests.

* Add tracking epoch times.

* fix transformer.

* Add examples_per_second.

* fix pylint

e0a2b8c3

15 Jul, 2019 1 commit
- Initial implementation of Shakespeare character LSTM. (#7218) · 395f6d2d
  Bruce Fontaine authored Jul 15, 2019
```
* Initial implementation of Shakespeare character LSTM.

* Fix import order
```
  395f6d2d