[ONNX] Update fuseLogSoftmaxNllLoss function to handle autocasting #51729
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
Contributor
💊 CI failures summary and remediationsAs of commit 8ae43ea (more details on the Dr. CI page):
🕵️ 4 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
BowenBao
reviewed
Feb 4, 2021
BowenBao
reviewed
Feb 4, 2021
Collaborator
|
would be great if you could add a test case with autocast |
added 2 commits
February 12, 2021 08:30
c1d1b3d to
aff0e0d
Compare
added 3 commits
February 12, 2021 08:39
BowenBao
approved these changes
Feb 17, 2021
BowenBao
pushed a commit
that referenced
this pull request
Feb 17, 2021
…51729) Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function [ghstack-poisoned]
This was referenced Feb 17, 2021
Closed
BowenBao
added a commit
that referenced
this pull request
Feb 18, 2021
…ocasting (#51729)" Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Differential Revision: [D26490326](https://our.internmc.facebook.com/intern/diff/D26490326) [ghstack-poisoned]
BowenBao
added a commit
that referenced
this pull request
Feb 19, 2021
…on to handle autocasting (#51729)" Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Differential Revision: [D26490326](https://our.internmc.facebook.com/intern/diff/D26490326) [ghstack-poisoned]
BowenBao
added a commit
that referenced
this pull request
Feb 19, 2021
…ocasting (#51729)" Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Differential Revision: [D26490326](https://our.internmc.facebook.com/intern/diff/D26490326) [ghstack-poisoned]
BowenBao
pushed a commit
to BowenBao/pytorch
that referenced
this pull request
Feb 19, 2021
…ytorch#51729) Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function
facebook-github-bot
pushed a commit
that referenced
this pull request
Feb 19, 2021
…51729) (#52349) Summary: Pull Request resolved: #52349 Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Test Plan: Imported from OSS Reviewed By: malfet Differential Revision: D26490326 Pulled By: SplitInfinity fbshipit-source-id: 4a6d806acc51b4696fd3932734d55af075fba6b1
iramazanli
pushed a commit
to iramazanli/pytorch
that referenced
this pull request
Feb 23, 2021
…ytorch#51729) (pytorch#52349) Summary: Pull Request resolved: pytorch#52349 Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Test Plan: Imported from OSS Reviewed By: malfet Differential Revision: D26490326 Pulled By: SplitInfinity fbshipit-source-id: 4a6d806acc51b4696fd3932734d55af075fba6b1
aocsa
pushed a commit
to Quansight/pytorch
that referenced
this pull request
Mar 15, 2021
…ytorch#51729) (pytorch#52349) Summary: Pull Request resolved: pytorch#52349 Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Test Plan: Imported from OSS Reviewed By: malfet Differential Revision: D26490326 Pulled By: SplitInfinity fbshipit-source-id: 4a6d806acc51b4696fd3932734d55af075fba6b1
xsacha
pushed a commit
to xsacha/pytorch
that referenced
this pull request
Mar 31, 2021
…ytorch#51729) (pytorch#52349) Summary: Pull Request resolved: pytorch#52349 Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss node and causing these patterns below not to be recognizable by peephole pass function Test Plan: Imported from OSS Reviewed By: malfet Differential Revision: D26490326 Pulled By: SplitInfinity fbshipit-source-id: 4a6d806acc51b4696fd3932734d55af075fba6b1
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
Adds a check for patterns for cases with autocasting enabled in which a cast node is inserted before the NegativeLogLikelihoodLoss
node and causing these patterns below not to be recognizable by peephole pass function