modelscope / modelscope/DiffSynth-Studio
Why was the pre-processing step of converting videos to latent representations removed?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.1k
- Forks
- 1.3k
- Avg merge
- 13h 12m
- Merged PRs (30d)
- 45
Description
The previous pipeline seemed to include an offline pre-processing step that converted raw videos into latent representations before training. This approach appeared to reduce training time by avoiding repeated encoding during each epoch.
Could you please clarify why this step was removed? Was it due to design changes, compatibility issues, or perhaps a shift toward end-to-end training?
Thanks in advance!
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
The issue names no file, test, or entry point. Compare the current video-training pipeline with the earlier preprocessing path and its history, then document the reason for the removal and any relevant compatibility or training-design constraints.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100