Skip to content

GitLab

"projects/vscode:/vscode.git/clone" did not exist on "601b44bfe00c8ab01b4bf4558fca2f68abe791b3"
Switch branch/tag
  1. 10 Mar, 2020 1 commit
    • Ran Chen's avatar
      Save to tmp directory on non-chief workers in model_training_utils · 682d36b5
      Ran Chen authored 
      In a multi worker set up saving is done on each worker. If they're saving to the same location, e.g. GCS, there will be conflicts. With this change we save to temporary directory on non-chief workers.

Note that, there may be synchronization in saving that needs all workers to participate, so we cannot only save on one worker.

PiperOrigin-RevId: 300141152
      682d36b5
  3. 07 Mar, 2020 1 commit
  5. 06 Mar, 2020 1 commit
    • Zongwei Zhou's avatar
      Temporarily disable explicit allreduce in BERT SQuAD · 11ccb99e
      Zongwei Zhou authored 
      In BERT SQuAD, disable explicit allreduce for now to keep the original clip_by_global_norm math. With explicit allreduce, the gradients before allreduce are scaled so even if we move clip_by_global_norm before allreduce (as in TF1 and pre-TF 2.2) it will operate on scaled gradients, the math will be changed. So with explicit allreduce, it is better to move clip_by_global_norm to after allreduce.

PiperOrigin-RevId: 299278082
      11ccb99e
  7. 05 Mar, 2020 1 commit
  9. 02 Mar, 2020 2 commits
  11. 26 Feb, 2020 1 commit
  13. 25 Feb, 2020 1 commit
  15. 20 Feb, 2020 1 commit
  17. 19 Feb, 2020 1 commit
  19. 13 Feb, 2020 1 commit
  21. 07 Feb, 2020 1 commit
  23. 23 Nov, 2019 1 commit
  25. 18 Nov, 2019 1 commit
  27. 28 Oct, 2019 1 commit
  29. 18 Oct, 2019 1 commit
  31. 11 Oct, 2019 1 commit
  33. 07 Oct, 2019 1 commit
  35. 24 Sep, 2019 2 commits
  37. 23 Sep, 2019 1 commit
  39. 17 Sep, 2019 1 commit
  41. 06 Sep, 2019 2 commits
  43. 04 Sep, 2019 1 commit
  45. 03 Sep, 2019 1 commit
  47. 16 Aug, 2019 1 commit
  49. 07 Aug, 2019 1 commit
  51. 06 Aug, 2019 1 commit
  53. 26 Jul, 2019 1 commit
    • Hongkun Yu's avatar
      Merged commit includes the following changes: (#7309) · 8c7a0e75
      Hongkun Yu authored 
      260060237  by zongweiz<zongweiz@google.com>:

    [BERT SQuAD] Enable mixed precision training

    Add mixed precision training support for BERT SQuAD model. Using the experimental Keras mixed precision API. For numeric stability, use fp32 for layer normalization, dense layers with GELU activation, etc.

--

PiperOrigin-RevId: 260060237
      8c7a0e75
  55. 25 Jul, 2019 1 commit
  57. 24 Jul, 2019 1 commit
  59. 19 Jul, 2019 1 commit
    • Jing Li's avatar
      Merged commit includes the following changes: (#7263) · c5a4978d
      Jing Li authored 
      * Merged commit includes the following changes:
258867180  by jingli<jingli@google.com>:

    Add new folders for upcoming reorg in model garden.

--
258893811  by hongkuny<hongkuny@google.com>:

    Adds summaries for metrics, allowing metrics inside keras.model.

--
258893048  by isaprykin<isaprykin@google.com>:

    Remove the `cloning` argument to `compile()`.

    Keras models are distributed by cloning in graph mode and without cloning in eager mode as of the change # 258652546.

--
258881002  by hongkuny<hongkuny@google.com>:

    Fix lint.

--
258874998  by hongkuny<hongkuny@google.com>:

    Internal

--
258872662  by hongkuny<hongkuny@google.com>:

    Fix doc

--

PiperOrigin-RevId: 258867180

* Create __init__.py

* Update __init__.py

* Update __init__.py

* Update __init__.py
      c5a4978d
  61. 16 Jul, 2019 1 commit
  63. 15 Jul, 2019 1 commit
  65. 24 Jun, 2019 1 commit
    • saberkun's avatar
      Merged commit includes the following changes: (#7093) · 240623ac
      saberkun authored 
      254785517  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Use train_single_step for BERT GPU models to temporarily work around some performance bugs in GPU runs

--
254497647  by hongkuny<hongkuny@google.com>:

    Fix device placement for TPU export model.

--

PiperOrigin-RevId: 254785517
      240623ac
  67. 20 Jun, 2019 1 commit
    • saberkun's avatar
      Merged commit includes the following changes: (#7060) · e0e6d981
      saberkun authored 
      254069984  by hongkuny<hongkuny@google.com>:
    Automated rollback of changelist 254060732.

254061429  by hongkuny<hongkuny@google.com>:

    Use host while loop for training steps.

--
254060732  by yifeif<yifeif@google.com>:
    Automated rollback of changelist 254027750.

254027750  by hongkuny<hongkuny@google.com>:

    Internal change

PiperOrigin-RevId: 254069984
      e0e6d981
  69. 18 Jun, 2019 1 commit
    • saberkun's avatar
      Merged commit includes the following changes: (#7049) · a1c47f28
      saberkun authored 
      253850824  by hongkuny<hongkuny@google.com>:

    Improve bert training utils.

--
253818191  by hongkuny<hongkuny@google.com>:

    Update savedmodel export to use new model.save() api.

--

PiperOrigin-RevId: 253850824
      a1c47f28
  71. 12 Jun, 2019 1 commit
    • David M. Chen's avatar
      Merged commit includes the following changes: (#6998) · ce03903f
      David M. Chen authored 
      252697519 by dmchen<dmchen@google.com>:

        BERT SQuAD accuracy test

25266352 by hongjunchoi<hongjunchoi@google.com>:

        Internal change

252647871 by hongjunchoi<hongjunchoi@google.com>:

        Enable multi worker TPU training for BERT pretraining.
      ce03903f
  73. 11 Jun, 2019 1 commit
    • saberkun's avatar
      Merged commit includes the following changes: (#6992) · f2eb1701
      saberkun authored 
      252522861  by hongkuny<hongkuny@google.com>:

    Remove export using trained model due to implementation error

--
252156812  by yuefengz<yuefengz@google.com>:

    Fix the callback method name in BERT: replaced on_batch_start with on_batch_begin. Without the fix, it won't work with Keras callbacks.

--
251782065  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

PiperOrigin-RevId: 252522861
      f2eb1701