- 04 Nov, 2021 1 commit
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 407461964
-
- 02 Nov, 2021 1 commit
-
-
Yuexin Wu authored
PiperOrigin-RevId: 407125012
-
- 26 Oct, 2021 1 commit
-
-
Terry Huang authored
PiperOrigin-RevId: 405693354
-
- 24 Sep, 2021 4 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 398790392
-
Hongkun Yu authored
PiperOrigin-RevId: 398790392
-
Rebecca Chen authored
PiperOrigin-RevId: 398612420
-
Rebecca Chen authored
PiperOrigin-RevId: 398612420
-
- 23 Sep, 2021 2 commits
-
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 398593113
-
Hongkun Yu authored
PiperOrigin-RevId: 398593113
-
- 22 Sep, 2021 2 commits
-
-
Frederick Liu authored
PiperOrigin-RevId: 398154690
-
Frederick Liu authored
PiperOrigin-RevId: 398154690
-
- 17 Sep, 2021 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 397244624
-
Hongkun Yu authored
PiperOrigin-RevId: 397244624
-
- 08 Aug, 2021 2 commits
- 05 Aug, 2021 4 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 389025611
-
Hongkun Yu authored
PiperOrigin-RevId: 389025611
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 389019529
-
A. Unique TensorFlower authored
PiperOrigin-RevId: 389019529
-
- 24 Jul, 2021 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 386654855
-
Hongkun Yu authored
PiperOrigin-RevId: 386654855
-
- 23 Jul, 2021 2 commits
-
-
Frederick Liu authored
PiperOrigin-RevId: 386531154
-
Frederick Liu authored
PiperOrigin-RevId: 386531154
-
- 21 Jul, 2021 2 commits
-
-
Frederick Liu authored
PiperOrigin-RevId: 386029066
-
Frederick Liu authored
PiperOrigin-RevId: 386029066
-
- 20 Jul, 2021 2 commits
- 15 Jul, 2021 2 commits
-
-
Hongkun Yu authored
PiperOrigin-RevId: 384957416
-
Hongkun Yu authored
PiperOrigin-RevId: 384957416
-
- 13 Jul, 2021 2 commits
- 09 Jul, 2021 2 commits
-
-
Reed Wanderman-Milne authored
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
-
Reed Wanderman-Milne authored
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
-
- 24 Jun, 2021 2 commits
-
-
A. Unique TensorFlower authored
Adds a feature to process a dictionary for `input_paths` in `DataConfig`, to allow combining multiple datasets using a user defined combine_fn. PiperOrigin-RevId: 381363688
-
A. Unique TensorFlower authored
Adds a feature to process a dictionary for `input_paths` in `DataConfig`, to allow combining multiple datasets using a user defined combine_fn. PiperOrigin-RevId: 381363688
-
- 23 Jun, 2021 2 commits
-
-
Reed Wanderman-Milne authored
In nlp/train.py and vision/beta/train.py, certain flags are marked as required. Additionally, in certain functions, error messages are improved if a necessary flag is not specified, which is a fallback in case a file calling define_flags() does not mark the necessary flags are required. Previously if any of these flags were not specified, it would crash with a cryptic error message, making it hard to tell what went wrong. In a subsequent change, I will mark flags as required in more files which call define_flags(). PiperOrigin-RevId: 381066985
-
Reed Wanderman-Milne authored
In nlp/train.py and vision/beta/train.py, certain flags are marked as required. Additionally, in certain functions, error messages are improved if a necessary flag is not specified, which is a fallback in case a file calling define_flags() does not mark the necessary flags are required. Previously if any of these flags were not specified, it would crash with a cryptic error message, making it hard to tell what went wrong. In a subsequent change, I will mark flags as required in more files which call define_flags(). PiperOrigin-RevId: 381066985
-
- 22 Jun, 2021 2 commits
- 20 Jun, 2021 1 commit
-
-
Abdullah Rashwan authored
PiperOrigin-RevId: 380470496
-