modelscope / modelscope/ms-swift
GRPO训练时，RLHF 进程无法连接到已启动的 rollout（vLLM server）
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
rollout命令行
rollout终端
rlhf命令行
rlhf终端
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 separate rollout and RLHF commands shown in the issue, then read the corresponding terminal output to identify where the connection fails. Compare the startup parameters and connection endpoints used by the rollout server and RLHF process. Done means the RLHF process connects successfully to the already running rollout server during GRPO training.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100