[DTensor] Exclude Partial placements for boolean tensors in validator#174538
Closed
wconstab wants to merge 5 commits intogh/wconstab/525/basefrom
Closed
[DTensor] Exclude Partial placements for boolean tensors in validator#174538wconstab wants to merge 5 commits intogh/wconstab/525/basefrom
wconstab wants to merge 5 commits intogh/wconstab/525/basefrom
Conversation
Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. Authored with Claude. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/174538
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 39 PendingAs of commit 4dce57c with merge base f365425 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Feb 8, 2026
…n validator" Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. Authored with Claude. [ghstack-poisoned]
pianpwk
reviewed
Feb 9, 2026
Contributor
pianpwk
left a comment
There was a problem hiding this comment.
would this be solved by banning all True/False boolean tensors? it seems like a nice rule to discover
…n validator" Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. Authored with Claude. [ghstack-poisoned]
…n validator" Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. Authored with Claude. [ghstack-poisoned]
…n validator" Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. 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
Partial decomposition creates float local values (with offsets for min/max, proportional splits for sum/avg) that lose boolean semantics when cast back to bool by ops like `where`. This caused a massive false positive explosion (134 "missing" rules for `where`) because any Partial placement on a True boolean scalar produced non-zero float values on all ranks, which all cast to True, making Partial equivalent to Replicate for that sample but not in general. Authored with Claude. ghstack-source-id: 3708ac5 Pull Request resolved: pytorch/pytorch#174538
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):
Partial decomposition creates float local values (with offsets for
min/max, proportional splits for sum/avg) that lose boolean semantics
when cast back to bool by ops like
where. This caused a massivefalse positive explosion (134 "missing" rules for
where) becauseany Partial placement on a True boolean scalar produced non-zero
float values on all ranks, which all cast to True, making Partial
equivalent to Replicate for that sample but not in general.
Authored with Claude.