modelscope / modelscope/ms-swift
swift全参数预训练minicpm -v 2.6时，进程一直在同一步卡死，所有显卡的CPU占用均为100%
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
Describe the bug
swift全参数预训练minicpm -v 2.6时，进程一直在同一步卡死，所有显卡利用率均为100%
Your hardware and system info
CUDA Version: 12.2
A800 8*40G显卡
torch==2.8.0
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
No file, test, command, or traceback is provided. Start by reproducing MiniCPM-V 2.6 full-parameter pretraining with the reported CUDA 12.2, A800, and torch 2.8.0 setup, then collect the stuck-step logs and identify the responsible training entry point. Done means the hang is explained and the affected setup completes that step or has a verified workaround.
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
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100