modelscope / modelscope/ms-swift
num_generations 可以 以时间换空间吗？显存不足支持大rollout TOT
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
想问能否修改num_generations的生成逻辑，比如过大的batch，可以拆分为多个小批次产生？训练速度慢点可以接受
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 locating the Python generation path that handles num_generations and rollout/TOT behavior; the issue names no files or tests. Compare the memory behavior of one large batch with smaller batches, and define done as supporting large rollouts within available GPU memory while preserving the requested generation behavior.
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
- Mostly clear
- Newbie friendliness
- 25/100