pytorch/torch/onnx
BowenBao 913f1f75b3 Revert "Revert [ONNX] Redesign inplace conversion" (#56675)
Summary:
Adjust how MutationRemover is used to avoid creating aliasDb multiple times for the same graph.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/56675

Reviewed By: pbelevich

Differential Revision: D27945692

Pulled By: SplitInfinity

fbshipit-source-id: a6c548438e88ddee18ef03a6f0461ab9eaaaa829
2021-04-22 22:22:16 -07:00
..
__init__.py Enforce PEP263 for PyTorch python codebase (#55346) 2021-04-06 18:31:38 -07:00
operators.py Fix typos, via a Levenshtein-type corrector (#31523) 2020-01-17 16:03:19 -08:00
symbolic_caffe2.py [quant][onnx] Support conversion of quantized sigmoid operator from pytorch to caffe2 (#34629) 2020-03-13 22:42:06 -07:00
symbolic_helper.py Add lint for unqualified type: ignore (#56290) 2021-04-21 08:07:23 -07:00
symbolic_opset7.py Warn about floor_divide performing incorrect rounding (#50281) (#50281) 2021-02-10 03:13:34 -08:00
symbolic_opset8.py Add lint for unqualified type: ignore (#56290) 2021-04-21 08:07:23 -07:00
symbolic_opset9.py Revert "Revert [ONNX] Redesign inplace conversion" (#56675) 2021-04-22 22:22:16 -07:00
symbolic_opset10.py Add lint for unqualified type: ignore (#56290) 2021-04-21 08:07:23 -07:00
symbolic_opset11.py Revert "Revert [ONNX] Redesign inplace conversion" (#56675) 2021-04-22 22:22:16 -07:00
symbolic_opset12.py [ONNX] Enable tensordot symbolic function. (#55654) (#56166) 2021-04-20 23:00:41 -07:00
symbolic_opset13.py [ONNX] Add repeat_interleave symbolic (#52855) (#53312) 2021-03-12 02:49:34 -08:00
symbolic_registry.py [ONNX] Support torch.isinf, torch.any and torch.all export to ONNX (#53328) (#53529) 2021-03-12 02:49:41 -08:00
utils.py Add standardOps match more input type in ORT (#53813) (#56172) 2021-04-21 17:58:08 -07:00