modelscope / modelscope/ms-swift
How to train Qwen3 VL embedding by Python?
Open
Nobody has claimed this yet.
question
stale
- 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 question or discussion topic. / 我已经搜索过现有的 issues，确认这是一个新的问题与讨论。
Question Description / 问题描述
背景：由于图片只能实时下载，不能直接适配jsonl格式，需要自定义dataset
如何使用python脚本完成 Qwen3 VL embedding 的训练？
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
The issue mentions only a Python script, a custom dataset for images downloaded at runtime, and Qwen3 VL embedding training; start by locating the Python training and dataset entry points in ms-swift. Document a reproducible workflow for the non-JSONL image data and define completion with a runnable example or documented command.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100