Bert Maher ac121165e2 Remove ReduceOp::accumulator (#52196) ... Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/52196 A reduction does not need to know the buffer into which its result will be written. This change gets us closer to being able to create reductions inside Compute, where we have access to the tensor axes. ghstack-source-id: 121813071 Test Plan: test_tensorexpr Reviewed By: ZolotukhinM Differential Revision: D26420107 Pulled By: bertmaher fbshipit-source-id: c8d8a99649adfd6de56fe53a728f5aa034a84f13		2021-02-17 23:36:23 -08:00
..
api	Fix autograd when inputs contains tensors without materialized grad_fn (#51940)	2021-02-11 09:22:15 -08:00
common
dist_autograd	Fix Windows build failure after DDP PR merged (#45335)	2020-09-25 12:37:50 -07:00
jit	Removes deprecated preprocess method from the backend interface (#52258)	2021-02-17 17:53:36 -08:00
rpc	Fix flaky TestTrainingLoop - TestE2ETensorPipe (#51939)	2021-02-10 22:14:06 -08:00
tensorexpr	Remove ReduceOp::accumulator (#52196)	2021-02-17 23:36:23 -08:00
__init__.py	remediation of S205607	2020-07-17 17:19:47 -07:00