Skip to content

GitLab

Switch branch/tag
  1. 19 Apr, 2021 1 commit
  3. 17 Apr, 2021 1 commit
  5. 16 Apr, 2021 1 commit
  7. 15 Apr, 2021 2 commits
  9. 14 Apr, 2021 1 commit
  11. 13 Apr, 2021 4 commits
    • Rebecca Chen's avatar
      Internal change · fbc1a8c2
      Rebecca Chen authored 
      PiperOrigin-RevId: 368157180
      fbc1a8c2
    • Jaehong Kim's avatar
      Internal change · 516f0402
      Jaehong Kim authored 
      PiperOrigin-RevId: 368130039
      516f0402
    • Fan Yang's avatar
      Add more pytype checking. · 615609b8
      Fan Yang authored 
      PiperOrigin-RevId: 368129317
      615609b8
    • Reed Wanderman-Milne's avatar
      Use nonexperimental mixed precision API. · 9a4d14a9
      Reed Wanderman-Milne authored 
      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
      9a4d14a9
  13. 12 Apr, 2021 4 commits
  15. 10 Apr, 2021 1 commit
    • Reed Wanderman-Milne's avatar
      Remove dynamic_loss_scale argument to define_performance. · 3803472a
      Reed Wanderman-Milne authored 
      All models which support loss scaling support dynamic loss scaling, so the argument has no purpose. It used to be that some models scaled the loss manually instead of using a LossScaleOptimizer, and so did not support dynamic loss scaling.

PiperOrigin-RevId: 367719521
      3803472a
  17. 09 Apr, 2021 1 commit
  19. 08 Apr, 2021 1 commit
  21. 06 Apr, 2021 3 commits
  23. 05 Apr, 2021 5 commits
  25. 03 Apr, 2021 1 commit
  27. 02 Apr, 2021 4 commits
  29. 31 Mar, 2021 2 commits
  31. 30 Mar, 2021 4 commits
  33. 29 Mar, 2021 4 commits