[DTensor] Handle NaN outputs in strategy validator#174539
Closed
wconstab wants to merge 5 commits intogh/wconstab/526/basefrom
Closed
[DTensor] Handle NaN outputs in strategy validator#174539wconstab wants to merge 5 commits intogh/wconstab/526/basefrom
wconstab wants to merge 5 commits intogh/wconstab/526/basefrom
Conversation
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. [ghstack-poisoned]
This was referenced Feb 8, 2026
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/174539
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 36 PendingAs of commit c48457f with merge base f365425 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Feb 8, 2026
wconstab
added a commit
that referenced
this pull request
Feb 8, 2026
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. ghstack-source-id: b4de8b9 Pull Request resolved: #174539
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. [ghstack-poisoned]
wconstab
added a commit
that referenced
this pull request
Feb 8, 2026
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. ghstack-source-id: d88dcb6 Pull Request resolved: #174539
pianpwk
approved these changes
Feb 9, 2026
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. [ghstack-poisoned]
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. [ghstack-poisoned]
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. [ghstack-poisoned]
Contributor
Author
|
squashed |
sandy-gags
pushed a commit
to sandy-gags/pytorch
that referenced
this pull request
Mar 12, 2026
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is considered valid. Also skip samples with all-NaN ground truth (like the existing all-zero skip) since NaN is invariant under all reduce ops. This fixes false positive "incorrect" reports for igamma/igammac, whose OpInfo samples include negative inputs that produce all-NaN outputs. Authored with Claude. ghstack-source-id: 0a11643 Pull Request resolved: pytorch/pytorch#174539
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
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.
Stack from ghstack (oldest at bottom):
Use equal_nan=True in torch.allclose comparison so that NaN == NaN is
considered valid. Also skip samples with all-NaN ground truth (like the
existing all-zero skip) since NaN is invariant under all reduce ops.
This fixes false positive "incorrect" reports for igamma/igammac, whose
OpInfo samples include negative inputs that produce all-NaN outputs.
Authored with Claude.