Update symbolics policy to emit aten::ATen for Caffe2 build only#74680
Update symbolics policy to emit aten::ATen for Caffe2 build only#74680thiagocrepaldi wants to merge 4 commits intopytorch:masterfrom
Conversation
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 97f380d (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
f8f3b7d to
86ce15c
Compare
86ce15c to
6bba258
Compare
984e567 to
3efb270
Compare
Signed-off-by: Thiago Crepaldi <thiago.crepaldi@microsoft.com>
3efb270 to
977a7da
Compare
|
@pytorchmergebot merge this |
|
Merge failed due to Matched rule superuser, but it was not reviewed yet by any of:glaringlee,anjali411,simpkins,bugra,dzhulgakov, ... |
|
@malfet adding you as we need to Also asking permission for |
|
@pytorchmergebot merge this |
) Summary: Currently ONNX exporter symbolics can emit ATen operators when `operator_export_type==ONNX_ATEN_FALLBACK`. However, this is a behavior specific to Caffe2 builds, as the intend use of `ONNX_ATEN_FALLBACK` is to emit ATen operators only when there is no ONNX equivalent. The reason Caffe2 choses to emit ATen operators when ONNX counterpart exists is for performance on their particular engine implementation, which might not be true for other implementations. e.g. ONNX Runtime can optimize the generated ONNX graph into something more efficient This PR must be merged only after #73954 Pull Request resolved: #74680 Approved by: https://github.com/garymm, https://github.com/malfet Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/eab3f428833ce9e11e2db97e516fe4b78ea2e4dc Reviewed By: seemethere Differential Revision: D35751465 fbshipit-source-id: 079d2a9db7174e68a9f8740f08364a598630dabc
Currently ONNX exporter symbolics can emit ATen operators when `operator_export_type==ONNX_ATEN_FALLBACK`. However, this is a behavior specific to Caffe2 builds, as the intend use of `ONNX_ATEN_FALLBACK` is to emit ATen operators only when there is no ONNX equivalent. The reason Caffe2 choses to emit ATen operators when ONNX counterpart exists is for performance on their particular engine implementation, which might not be true for other implementations. e.g. ONNX Runtime can optimize the generated ONNX graph into something more efficient This PR must be merged only after #73954 Pull Request resolved: #74680 Approved by: https://github.com/garymm, https://github.com/malfet (cherry picked from commit eab3f42)
Currently ONNX exporter symbolics can emit ATen operators when
operator_export_type==ONNX_ATEN_FALLBACK. However, this is a behavior specific to Caffe2 builds, as the intend use ofONNX_ATEN_FALLBACKis to emit ATen operators only when there is no ONNX equivalent.The reason Caffe2 choses to emit ATen operators when ONNX counterpart exists is for performance on their particular engine implementation, which might not be true for other implementations. e.g. ONNX Runtime can optimize the generated ONNX graph into something more efficient
This PR must be merged only after #73954