pytorch/test/cpp
Nick Gibson 6fb5ce5569 [NNC] Fix some bugs in Round+Mod simplification (#42934)
Summary:
When working on the Cuda Codegen, I found that running the IRSimplifier before generating code lead to test fails. This was due to a bug in Round+Mod simplification (e.g. (x / y * y) + (x % y) => x) to do with the order in which the terms appeared. After fixing it and writing a few tests around those cases, I found another bug in simplification of the same pattern and have fixed it (with some more test coverage).

Pull Request resolved: https://github.com/pytorch/pytorch/pull/42934

Reviewed By: zhangguanheng66

Differential Revision: D23085548

Pulled By: nickgg

fbshipit-source-id: e780967dcaa7a5fda9f6d7d19a6b7e7b4e94374b
2020-08-13 09:47:21 -07:00
..
api Don't materialize output grads (#41821) 2020-08-11 04:27:07 -07:00
common
dist_autograd Release DistAutogradContainer context for each dist_autograd test case (#38711) 2020-06-18 20:58:55 -07:00
jit Fix lite trainer unit test submodule registration (#42714) 2020-08-07 18:26:56 -07:00
rpc Enroll TensorPipe agent in C++-only E2E test (#42680) 2020-08-13 07:07:30 -07:00
tensorexpr [NNC] Fix some bugs in Round+Mod simplification (#42934) 2020-08-13 09:47:21 -07:00
__init__.py remediation of S205607 2020-07-17 17:19:47 -07:00