modelscope / modelscope/DiffSynth-Studio
Question about qwen-image-edit training (resize)
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.1k
- Forks
- 1.3k
- Avg merge
- 13h 12m
- Merged PRs (30d)
- 45
Description
-
When training Qwen Image Edit, if edit_image_auto_resize is enabled but the generated (target) images are not resized accordingly, will this cause any mismatch?
-
During training, the method pipe.check_resize_height_width modifies the image resolution (height and width), but the latent tensor remains at its original size. This lead to dimensional inconsistencies
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by tracing the Qwen Image Edit training path that uses edit_image_auto_resize and pipe.check_resize_height_width, then compare the resized image dimensions with the latent tensor dimensions. Confirm whether the reported mismatch occurs and document the expected behavior or reproduction details; the issue does not name specific files or tests.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100