modelscope / modelscope/ms-swift
[RFC] Support Megatron SimPO Training
Open
Nobody has claimed this yet.
enhancement
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
rlhf.py: error: argument --rlhf_type/--rlhf-type: invalid choice: 'simpo' (choose from 'dpo', 'kto', 'rm')
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 in rlhf.py at the --rlhf_type/--rlhf-type argument, where the current choices are dpo, kto, and rm. Read the surrounding RLHF training entry points and determine what SimPO support requires; done means the command accepts simpo and supports Megatron SimPO training.
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
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100