modelscope / modelscope/ms-swift
DeepseekOCR全参数微调使用Megatron-swift显示还不支持
Open
Nobody has claimed this yet.
enhancement
more models
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
目前使用swift sft可以跑通deepseekOCR的全参微调，但速度太慢了。尝试使用megatron-swift进行全参微调，但报错：AssertionError: Model: /home/xx/mnt-data/open_source/DeepSeek-OCR/modelfile4sft/ is not supported.显示还不支持
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 the DeepSeekOCR full-parameter fine-tuning attempt with megatron-swift and inspect the model-support check that raises the reported AssertionError. Trace how the model path is recognized, then verify that the same training workflow completes without the unsupported-model error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100