pytorch

f35e0696 - Back out "Make grad point to bucket buffer in DDP to save memory usage" (#43557)

Back out "Make grad point to bucket buffer in DDP to save memory usage" (#43557) Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/43557 backout the diff that caused some errors in pytext distributed training Test Plan: Tested by rayhou who verified reverting the diff works Differential Revision: D23320238 fbshipit-source-id: caa0fe74404059e336cd95fdb41373f58ecf486e