modelscope / modelscope/ms-swift
能实现 transformers 版本的 MTP 训练吗？
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
Checklist / 检查清单
- I have searched existing issues, and this is a new feature request. / 我已经搜索过现有的 issues，确认这是一个新的 Feature Request。
Feature Request Description / Feature Request 描述
Megatron-LLM 的依赖过于复杂，在服务器上安装不上去。
请问可以实现一个 transformers 的吗？直接在 swift sft 里面使用 mtp loss。
Pull Request / Pull Request 信息
No response
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 tracing the existing Swift SFT entry point and its Megatron-LLM integration, then identify where MTP loss is handled. Compare that path with the requested Transformers-based training flow and define the supported model and loss boundaries. Done means Swift SFT can run MTP loss without requiring the Megatron-LLM dependency.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100