pytorch/torch/distributed/algorithms
Yi Wang fccfe7bd1a [Gradient Compression] Add unit tests that test default Python comm hook implementations (#47158)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/47158

1. Test the default Python comm hook implementations ALLREDUCE and FP16_COMPRESS, besides an ad-hoc all-reduce implementation.
2. Typo fix.
3. Reformat default_hooks.py.
4. Publish register_comm_hook API for DDP module (This should be done in a separate diff, but got merged unintentionally.)

The new style can be used for testing any new comm hook like PowerSGD easily.
Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202

ghstack-source-id: 116012600

Test Plan: buck test mode/dev-nosan caffe2/test/distributed:c10d -- test_default_ddp_comm_hooks_nccl

Reviewed By: rohan-varma

Differential Revision: D24669639

fbshipit-source-id: 048c87084234edc2398f0ea6f01f2f083a707939
2020-11-06 00:28:09 -08:00
..
ddp_comm_hooks [Gradient Compression] Add unit tests that test default Python comm hook implementations (#47158) 2020-11-06 00:28:09 -08:00
__init__.py [Gradient Compression] Add unit tests that test default Python comm hook implementations (#47158) 2020-11-06 00:28:09 -08:00