Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Horovod: fixed early stopping and added metrics aggregation #3775

Merged
merged 42 commits into from
Nov 5, 2020
Merged

Horovod: fixed early stopping and added metrics aggregation #3775

merged 42 commits into from
Nov 5, 2020

Commits on Oct 9, 2020

  1. Fixed early stopping for Horovod

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    1c1a83e View commit details
    Browse the repository at this point in the history

  2. Refactored to sync_dist_if_available

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    f808d33 View commit details
    Browse the repository at this point in the history

  3. Bump min Horovod version to support hvd.is_initialized

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    7774093 View commit details
    Browse the repository at this point in the history

  4. Changelog

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    fd6fb80 View commit details
    Browse the repository at this point in the history

  5. Added back change for Horovod

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    239fcd6 View commit details
    Browse the repository at this point in the history

  6. Removed redundant checks for initialization

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    660651c View commit details
    Browse the repository at this point in the history

  7. Implement metrics gathering for Horovod

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    f06d7e4 View commit details
    Browse the repository at this point in the history

  8. Added test for EvalResult

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    49f583d View commit details
    Browse the repository at this point in the history

  9. Renamed ddp_sync_on_step -> dist_sync_on_step

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    3da46e7 View commit details
    Browse the repository at this point in the history

  10. Added metric test for Horovod

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    4f770d4 View commit details
    Browse the repository at this point in the history

  11. Added option pass callable allgather function to metric base class

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    905bdd1 View commit details
    Browse the repository at this point in the history

  12. Added dist_sync_fn

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    39cedfc View commit details
    Browse the repository at this point in the history

  13. Fixed calls to private _sync_dist

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    5818466 View commit details
    Browse the repository at this point in the history

  14. Fixed Horovod test

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    857ab82 View commit details
    Browse the repository at this point in the history

  15. Added sync_tensor to the distributed backend

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    98d5325 View commit details
    Browse the repository at this point in the history

  16. Skip Windows

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    1596f93 View commit details
    Browse the repository at this point in the history

  17. Insert test path

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    2350ba0 View commit details
    Browse the repository at this point in the history

  18. Removed redundant import

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    5f08689 View commit details
    Browse the repository at this point in the history

  19. Updated drone

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    4e3ce48 View commit details
    Browse the repository at this point in the history

  20. Unset HOROVOD_GPU_ALLREDUCE

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    1c88684 View commit details
    Browse the repository at this point in the history

  21. Unset

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    825278c View commit details
    Browse the repository at this point in the history

  22. No cache dir

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    ae45024 View commit details
    Browse the repository at this point in the history

  23. No uninstall

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    a4ef8d0 View commit details
    Browse the repository at this point in the history

  24. Unset variables

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    4aef1a1 View commit details
    Browse the repository at this point in the history

  25. Uninstall Horovod during initialization

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    c42666f View commit details
    Browse the repository at this point in the history

  26. Replaced more references to ddp_sync_on_step

    @tgaddair
    tgaddair committed Oct 9, 2020
    Configuration menu
    Copy the full SHA
    46d0732 View commit details
    Browse the repository at this point in the history

Commits on Oct 22, 2020

  1. Merge branch 'master' of https://github.com/PyTorchLightning/pytorch-… 

    …lightning into hvd_early_stop
    @teddykoker
    teddykoker committed Oct 22, 2020
    Configuration menu
    Copy the full SHA
    b416730 View commit details
    Browse the repository at this point in the history

Commits on Nov 3, 2020

  1. Resolved merge conflicts

    @tgaddair
    tgaddair committed Nov 3, 2020
    Configuration menu
    Copy the full SHA
    0326b2a View commit details
    Browse the repository at this point in the history

  2. Fixed imports

    @tgaddair
    tgaddair committed Nov 3, 2020
    Configuration menu
    Copy the full SHA
    ab8dd34 View commit details
    Browse the repository at this point in the history

  3. Fixed attribute

    @tgaddair
    tgaddair committed Nov 3, 2020
    Configuration menu
    Copy the full SHA
    3590051 View commit details
    Browse the repository at this point in the history

  4. Added back default

    @tgaddair
    tgaddair committed Nov 3, 2020
    Configuration menu
    Copy the full SHA
    80532eb View commit details
    Browse the repository at this point in the history

Commits on Nov 4, 2020

  1. Merge branch 'master' into hvd_early_stop

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    d7409ea View commit details
    Browse the repository at this point in the history

  2. Lint

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    2a00e1a View commit details
    Browse the repository at this point in the history

  3. Merge branch 'master' into hvd_early_stop

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    0fd1e9b View commit details
    Browse the repository at this point in the history

  4. Added back docstring

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    603724a View commit details
    Browse the repository at this point in the history

  5. Made gather_all_tensors default

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    bafdb9d View commit details
    Browse the repository at this point in the history

  6. Added whitespace

    @tgaddair
    tgaddair committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    1177a96 View commit details
    Browse the repository at this point in the history

  7. Merge branch 'master' into hvd_early_stop

    @SeanNaren
    SeanNaren committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    63d8bdb View commit details
    Browse the repository at this point in the history

  8. Update tests/models/test_horovod.py 

    Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
    @tgaddair @Borda
    tgaddair and Borda committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    2b19746 View commit details
    Browse the repository at this point in the history

  9. Update pytorch_lightning/metrics/metric.py 

    Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
    @tgaddair @Borda
    tgaddair and Borda committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    77ee13f View commit details
    Browse the repository at this point in the history

  10. Update CHANGELOG.md

    @Borda
    Borda committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    855fab7 View commit details
    Browse the repository at this point in the history

  11. Merge branch 'master' into hvd_early_stop

    @SeanNaren
    SeanNaren committed Nov 4, 2020
    Configuration menu
    Copy the full SHA
    8b1ee77 View commit details
    Browse the repository at this point in the history