Fix corrupt image handling in custom dataset#126
Open
sushantkhemalapure wants to merge 7 commits intohumanai-foundation:mainfrom
Open
Fix corrupt image handling in custom dataset#126sushantkhemalapure wants to merge 7 commits intohumanai-foundation:mainfrom
sushantkhemalapure wants to merge 7 commits intohumanai-foundation:mainfrom
Conversation
- Updated TrOCR model path from ../../weights to ../../models - Ensures correct model loading instead of CRAFT weights
- Corrected unpacking of image.shape[:2] to (height, width) - Prevents distortion in rotated images
- Added sys.path handling to locate CRAFT module - Ensures qapp.py runs consistently outside Docker
- Prevent returning invalid sample types - Raise error or safely skip unreadable images
Author
|
Handled invalid image cases to avoid breaking the DataLoader. Let me know if you'd prefer a different strategy (skip vs error). |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Issue
In
custom_dataset.py, when an image cannot be read (e.g., corrupt or missing), the dataset returns an unexpected type (such as a string or tuple) instead of the expected sample dictionary. This can break the DataLoader during training or evaluation.Fix
Updated the dataset logic to handle unreadable images safely by ensuring only valid sample formats are returned. Invalid samples are either skipped or handled with a clear error.
Impact
Testing