Skip to content

GitLab

Switch branch/tag
  1. 19 Nov, 2021 1 commit
  3. 17 Nov, 2021 1 commit
  5. 04 Mar, 2021 1 commit
  7. 22 Sep, 2020 1 commit
  9. 26 Aug, 2020 4 commits
  11. 25 Aug, 2020 1 commit
  13. 10 Aug, 2020 1 commit
  15. 23 Jul, 2020 2 commits
  17. 27 May, 2020 1 commit
  19. 04 May, 2020 1 commit
  21. 21 Apr, 2020 1 commit
  23. 20 Apr, 2020 1 commit
  25. 17 Apr, 2020 1 commit
  27. 16 Apr, 2020 1 commit
  29. 14 Apr, 2020 2 commits
  31. 12 Apr, 2020 1 commit
  33. 04 Apr, 2020 2 commits
  35. 01 Apr, 2020 1 commit
  37. 16 Mar, 2020 1 commit
  39. 13 Mar, 2020 2 commits
  41. 12 Mar, 2020 1 commit
  43. 24 Feb, 2020 1 commit
  45. 23 Jan, 2020 1 commit
  47. 19 Dec, 2019 1 commit
  49. 05 Dec, 2019 1 commit
    • Haoyu Zhang's avatar
      Remove tweaked benchmark for no_dist_strat in function mode. · ef8aed79
      Haoyu Zhang authored 
      No need to test explicit_gpu_placement + no_dist_strat + function case, since the performance issue is already fixed by unifying the Keras codepaths. However, there is still performance gap using the explicit_gpu_placement flag when run_eagerly.

PiperOrigin-RevId: 283890008
      ef8aed79
  51. 27 Nov, 2019 1 commit
  53. 25 Nov, 2019 1 commit
    • Sai Ganesh Bandiatmakuri's avatar
      Inject enable_runtime_flags into benchmarks. · bcce419a
      Sai Ganesh Bandiatmakuri authored 
      This will help general debugging by enabling custom execution with  --benchmark_method_steps.

E.g --benchmark_method_steps=train_steps=7 will run the benchmark for only 7 steps without modifying benchmark code.

PiperOrigin-RevId: 282396875
      bcce419a
  55. 06 Nov, 2019 1 commit
  57. 28 Oct, 2019 1 commit
  59. 19 Sep, 2019 2 commits
  61. 13 Sep, 2019 1 commit
  63. 12 Sep, 2019 1 commit