1. 11 Mar, 2022 1 commit
  2. 18 Jan, 2022 1 commit
  3. 03 Dec, 2021 1 commit
  4. 22 Nov, 2021 1 commit
  5. 12 Sep, 2021 1 commit
  6. 03 Aug, 2021 1 commit
  7. 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
  8. 03 Mar, 2021 1 commit
  9. 22 Jan, 2021 1 commit
  10. 05 Jan, 2021 1 commit
  11. 02 Jan, 2021 1 commit
  12. 18 Dec, 2020 1 commit
  13. 13 Nov, 2020 1 commit
  14. 17 Sep, 2020 1 commit
  15. 13 Sep, 2020 1 commit
  16. 01 Sep, 2020 1 commit