Fix NonZero eager impl. #12143
Merged
Fix NonZero eager impl. #12143
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WilBrady
commented
Jul 11, 2022
WilBrady
force-pushed
the
wilbrady/fixNonzero
branch
2 times, most recently
from
4e94c33 to
e81d936
Compare
msftlincoln
reviewed
Jul 12, 2022
WilBrady
force-pushed
the
wilbrady/fixNonzero
branch
from
e81d936 to
ee150d7
Compare
msftlincoln
previously approved these changes
Jul 12, 2022
msftlincoln
previously approved these changes
Jul 12, 2022
msftlincoln
approved these changes
Jul 12, 2022
|
This pull request introduces 4 alerts when merging d99e80f into a6fd1a3 - view on LGTM.com new alerts:
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description: Nonzero is not a typical unary op as the output shape is not the same as the input self tensor.
Motivation and Context
The current nonzero implementation in eager mode is not correctly aligned with aten behavior. First Onnx produces an output that is transposed to that of aten and pytorch. Second the output is a list of indexes and not the same shape as the input tensor. For those reasons, we need to directly implement both nonzero and nonzero.out.