Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ms-swift 3.1版本中多轮对话+单轮grounding(多轮对话)数据集构建 #3088

Open
corkiyao opened this issue Feb 13, 2025 · 0 comments

Comments

@corkiyao
Copy link

corkiyao commented Feb 13, 2025

在自定义数据集的时候,我不清楚ms-swift3.1版本的多轮对话grounding数据集格式。我只知道单轮对话。因为第一次使用,这个部分不太清楚,期待作者可以回复,非常感谢。
具体来说,比如:

这是监督微调的数据集格式:
{"messages": [{"role": "system", "content": "<system>"}, {"role": "user", "content": "<query1>"}, {"role": "assistant", "content": "<response1>"}, {"role": "user", "content": "<query2>"}, {"role": "assistant", "content": "<response2>"}]}

我注意到这是多轮对话的格式,另外下面的是目标定位的格式:

{"messages": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "<image>找到图像中的<ref-object>"}, {"role": "assistant", "content": "<bbox><bbox>"}], "images": ["/xxx/x.jpg"], "objects": {"ref": ["羊"], "bbox": [[90.9, 160.8, 135, 212.8], [360.9, 480.8, 495, 532.8]]}}

但是我想先使用图像描述作为第一轮,之后定位作为第二轮。但是按照作者提供的格式,似乎不满足这样的方式。

请问可以这样吗?

{"messages": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "<image>请问这是一个青青草原吗?"}, {"role": "assistant", "content": "是滴,这是一个青青草原。"}, {"role": "user", "content": "<image>找到图像中的<ref-object>"}, {"role": "assistant", "content": "<bbox><bbox>"}], "images": ["/xxx/x.jpg"], "objects": {"ref": ["羊"], "bbox": [[90.9, 160.8, 135, 212.8], [360.9, 480.8, 495, 532.8]]}}

先构建一个图像描述,再增加目标定位。?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant