1. 04 Jan, 2022 1 commit
  2. 29 Dec, 2021 1 commit
  3. 28 Dec, 2021 1 commit
  4. 01 Dec, 2021 1 commit
  5. 29 Nov, 2021 1 commit
  6. 22 Nov, 2021 2 commits
  7. 19 Oct, 2021 3 commits
  8. 12 Oct, 2021 1 commit
  9. 09 Oct, 2021 1 commit
  10. 07 Oct, 2021 1 commit
  11. 02 Oct, 2021 1 commit
  12. 22 Sep, 2021 1 commit
  13. 15 Sep, 2021 1 commit
  14. 12 Sep, 2021 1 commit
  15. 09 Sep, 2021 1 commit
  16. 07 Sep, 2021 1 commit
  17. 31 Aug, 2021 1 commit
  18. 27 Aug, 2021 1 commit
  19. 25 Aug, 2021 1 commit
  20. 03 Aug, 2021 1 commit
  21. 06 Jun, 2021 2 commits
  22. 02 Jun, 2021 1 commit
  23. 27 May, 2021 1 commit
  24. 24 May, 2021 1 commit
  25. 17 May, 2021 2 commits
  26. 11 May, 2021 1 commit
  27. 26 Apr, 2021 1 commit
  28. 13 Apr, 2021 1 commit
  29. 12 Apr, 2021 1 commit
    • Reed Wanderman-Milne's avatar
      Use nonexperimental mixed precision API for official models. · 0d8f9807
      Reed Wanderman-Milne authored
      For all modified calls to set_mixed_precision_policy(), the loss_scale argument was removed, as it cannot be passed if the nonexperimental API is used. For all such callers, the loss_scale is later used to explicitly create a LossScaleOptimizer, so removing the argument has no impact.
      
      Switching to the non-experimental LossScaleOptimizer has no effect, as it has near identical behavior and all isinstance checks within the official models check for the non-experimental version.
      
      PiperOrigin-RevId: 368101975
      0d8f9807
  30. 08 Apr, 2021 1 commit
  31. 07 Apr, 2021 1 commit
  32. 01 Apr, 2021 2 commits
  33. 10 Mar, 2021 1 commit
  34. 25 Jan, 2021 1 commit