pytorch/torch/distributed/algorithms
Yi Wang b03b45afd9 [DDP Comm Hook] Use a single tensor instead of a tensor list as the comm hook result (#62074)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/62074

Since SPMD mode is retired, the comm hook result will always be a single tensor.

This can improve comm hook developer experience, as no need to add an extra `[0]` to the precursor future result.

#Closes: https://github.com/pytorch/pytorch/issues/61914
ghstack-source-id: 134164593

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

Reviewed By: rohan-varma

Differential Revision: D29864732

fbshipit-source-id: 59fe6dd78b66214b1788514ad4d236039d9bda31
2021-07-23 03:32:05 -07:00
..
ddp_comm_hooks [DDP Comm Hook] Use a single tensor instead of a tensor list as the comm hook result (#62074) 2021-07-23 03:32:05 -07:00
model_averaging [Model Averaging] Skip model averaging for the first K steps (#61207) 2021-07-10 17:12:16 -07:00
__init__.py [Gradient Compression] Add unit tests that test default Python comm hook implementations (#47158) 2020-11-06 00:28:09 -08:00
join.py Minor documentation fixes (#61785) 2021-07-19 09:01:29 -07:00