Fixing Linear Semantic Segmentation Bugs - #880
Merged
Merged
Conversation
mrpositron
changed the title
mrpositron
marked this pull request as ready for review
July 20, 2026 08:45
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 20f45e5a10
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Contributor
Author
|
/review |
yutong-xiang-97
approved these changes
Jul 21, 2026
mrpositron
enabled auto-merge (squash)
July 21, 2026 14:42
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What has changed and why?
Fixes two bugs in linear semantic segmentation with DINOv3 backbones:
DINOv3 builders hardcode
drop_path_rateinternally and re-splat**kwargs, so forwarding it raisedgot multiple values for keyword argument 'drop_path_rate'. No DINOv3 builder accepts the argument, sodinov3_package.pynow drops it before building the model (a no-op for the model).Validation previously tiled every image and forwarded all generated crops in a single batch while retaining every crop’s full-resolution logits. Because the number of crops grows with the image aspect ratio, memory usage was unbounded for very wide or tall images.
Validation now:
viz_max_imagespredictions on CPU for visualization.This bounds forward activation and crop-logit memory while preserving correct reconstruction across overlapping crops.
How has it been tested?
test_dinov3_package.pyassertingdrop_path_rateis dropped and the model builds.test_train_model.pycovering the per-image validation path (loss/metrics accumulation, chunked forward, andviz_max_imagescapping).Did you update CHANGELOG.md?
Did you update the documentation?