Fix affine() IndexError with boolean mask indexing#1547
Merged
alinpahontu2912 merged 1 commit intodotnet:mainfrom Mar 3, 2026
Merged
Fix affine() IndexError with boolean mask indexing#1547alinpahontu2912 merged 1 commit intodotnet:mainfrom
alinpahontu2912 merged 1 commit intodotnet:mainfrom
Conversation
Replace img[mask] = fill_img[mask] with torch::where(mask, fill_img, img) in THSVision_ApplyGridTransform's Nearest-mode fill logic. The operator[] boolean mask indexing is not supported in recent LibTorch versions. This fixes torchvision.transforms.functional.affine() and rotate() when called with a fill value and Nearest interpolation mode. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
alinpahontu2912
changed the title
Contributor
There was a problem hiding this comment.
Pull request overview
Fixes a regression in TorchVision-style affine/rotate transforms with fill by removing unsupported boolean-mask operator[] indexing in the native grid-transform path.
Changes:
- Replace
img[mask] = fill_img[mask]withtorch::where(mask, fill_img, img)inTHSVision_ApplyGridTransformfor Nearest-mode fill handling. - Add regression tests covering affine transforms on 3D/4D inputs (including batched 4D) and rotate with fill.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
src/Native/LibTorchSharp/THSVision.cpp |
Avoids unsupported boolean mask indexing by using torch::where to apply fill under Nearest sampling. |
test/TorchSharpTest/TestTorchVision.cs |
Adds tests to ensure affine/rotate with fill execute successfully for common input shapes. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
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 #1502
Replace img[mask] = fill_img[mask] with torch::where(mask, fill_img, img) in THSVision_ApplyGridTransform's Nearest-mode fill logic. The operator[] boolean mask indexing is not supported in recent LibTorch versions.
This fixes torchvision.transforms.functional.affine() and rotate() when called with a fill value and Nearest interpolation mode.