8000 label问题 · Issue #40 · tulerfeng/Video-R1 · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

label问题 #40

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.

Sign up for GitHub

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

Jump to bottom
Open
Graysonicc opened this issue Apr 25, 2025 · 1 comment
Open

label问题 #40

Graysonicc opened this issue Apr 25, 2025 · 1 comment

Comments

@Graysonicc
Copy link
Graysonicc commented Apr 25, 2025 •
edited
Loading

SFT时候，请问你们是将整个question加answer都作为label吗（除image和video pad部分），我看源码中是这样写的。但是常见的不应该是将整个question部分都-100掩码住，只用answer部分参与计算loss吗？
@tulerfeng
Copy link
Owner
tulerfeng commented Apr 25, 2025 •
edited
Loading

您好，我们主要是参考R1-V和TRL里面的代码来写的，我目前也不太清楚为啥会是这样，参考代码如下：

https://github.com/huggingface/trl/blob/main/examples/scripts/sft_video_llm.py

https://github.com/Deep-Agent/R1-V/blob/main/src/r1-v/src/open_r1/sft.py

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@tulerfeng @Graysonicc
0