modelscope / modelscope/ms-swift
qwen3omni Loading checkpoint 太慢
Open
Nobody has claimed this yet.
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
使用swift训练的时候 loading checkpoint 需要30分钟，直接使用transforms 加载很快。
环境：swift 3.10镜像
Contributor guide
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 reproducing checkpoint loading during swift training in the swift 3.10 image, then compare its behavior and timing with loading the same checkpoint directly through transformers. Confirm the model and checkpoint used for qwen3omni, identify where the extra delay occurs, and consider the issue done when swift training loads it without the reported 30-minute delay.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 38/100