- 11 Oct, 2019 2 commits
-
-
Hongkun Yu authored
* Revert "Update tf.contrib.data to tf.data.experimental. (#7650)" This reverts commit faf4bbb3. * revert research
-
Derek Murray authored
-
- 10 Oct, 2019 2 commits
-
-
A. Unique TensorFlower authored
change benchmark's log verbosity to logging.INFO. it seems to me that DEBUG map to ---v=1 internally, which is way to verbose for the purpose of benchmarking. PiperOrigin-RevId: 274040907
-
Hongkun Yu authored
PiperOrigin-RevId: 273966871
-
- 09 Oct, 2019 1 commit
-
-
Reed Wanderman-Milne authored
Instead of needing to ensure variables are float32, casting inputs to float32, etc, instead dtype="float32" is passed to the layer constructor, which will do all that logic automatically. The only difference is the output of LayerNorm is now float32 instead of float16, so an extra cast is needed elsewhere. PiperOrigin-RevId: 273833286
-
- 07 Oct, 2019 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 273371605
-
- 18 Sep, 2019 1 commit
-
-
David Chen authored
PiperOrigin-RevId: 269914707
-
- 17 Sep, 2019 1 commit
-
-
Hongkun Yu authored
Refactor basic utils to modeling/ PiperOrigin-RevId: 269600561
-
- 16 Sep, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 269376599
-
- 09 Sep, 2019 2 commits
-
-
David Chen authored
PiperOrigin-RevId: 268057019
-
Reed Wanderman-Milne authored
--stop_threshold, --num_gpu, --hooks, --export_dir, and --distribution_strategy have been unexposed from models which do not use them PiperOrigin-RevId: 268032080
-
- 05 Sep, 2019 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 267435985
-
- 04 Sep, 2019 1 commit
-
-
Reed Wanderman-Milne authored
--clean, --train_epochs, and --epochs_between_evals have been unexposed from models which do not use them PiperOrigin-RevId: 267065651
-
- 03 Sep, 2019 1 commit
-
-
Gaurav Jain authored
PiperOrigin-RevId: 266848625
-
- 30 Aug, 2019 2 commits
-
-
Reed Wanderman-Milne authored
I soon plan on making mixed precision policies exposed in TF2 only. PiperOrigin-RevId: 266433372
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 266413847
-
- 29 Aug, 2019 1 commit
-
-
Reed Wanderman-Milne authored
The transformer model requires TF2. PiperOrigin-RevId: 266026991
-
- 28 Aug, 2019 2 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 265994446
-
Adrian Kuegel authored
PiperOrigin-RevId: 265856020
-
- 27 Aug, 2019 1 commit
-
-
Hongkun Yu authored
PiperOrigin-RevId: 265771605
-
- 26 Aug, 2019 1 commit
-
-
Igor Saprykin authored
PiperOrigin-RevId: 265499806
-
- 23 Aug, 2019 2 commits
-
-
David Chen authored
PiperOrigin-RevId: 265122309
-
David Chen authored
PiperOrigin-RevId: 264958330
-
- 22 Aug, 2019 3 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 264867908
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 264853703
-
Adrian Kuegel authored
PiperOrigin-RevId: 264776396
-
- 21 Aug, 2019 2 commits
-
-
Reed authored
-
Vinh Nguyen authored
-
- 20 Aug, 2019 3 commits
-
-
Reed authored
The old infer_float32_policies policy will be removed from TensorFlow soon.
-
Vinh Nguyen authored
-
Hongkun Yu authored
PiperOrigin-RevId: 264300408
-
- 19 Aug, 2019 3 commits
-
-
Reed Wanderman-Milne authored
An earlier change made it so running this file would cause an error. The unit tests still passed, as the unit tests do not directly call the main() function PiperOrigin-RevId: 264268400
-
Toby Boyd authored
Small tweak to the range based on seeing values slightly over 29 and slightly under 28. This avoids false positives that even if valid are not actionable.
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 264174941
-
- 16 Aug, 2019 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 263863438
-
Reed Wanderman-Milne authored
PiperOrigin-RevId: 263812316
-
- 15 Aug, 2019 3 commits
-
-
Igor Saprykin authored
PiperOrigin-RevId: 263669433
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 263485329
-
Igor Saprykin authored
PiperOrigin-RevId: 263463300
-
- 09 Aug, 2019 1 commit
-
-
Vinh Nguyen authored
-