Repository navigation
Conversation
Signed-off-by: 0z5a <dezhen.lu@student.uni-tuebingen.de>
pranavm-nvidia
approved these changes
Nov 5, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #4916.
Widen INT32 outputs to INT64 before computing differences so the simple comparator does not wrap and falsely pass at zero tolerance. For PyTorch INT64 arrays, compute the absolute value directly rather than round-tripping through FLOAT32, which also loses low integer bits.
Testing: the new INT32 boundary regressions fail on the original implementation. On CPU, 132 tests passed and 4 were skipped across the existing simple/L2 comparator tests and the new NumPy/PyTorch tests. Runtime warnings from division by zero and non-finite statistics were present in the test run. Wider INT64 subtraction overflow is outside this change's scope. No GPU or TensorRT sample tests were run.
This draft is provided for the issue discussion; it can be marked ready after the issue has been reviewed as required by the contribution guidelines.