Skip to content

fix qwen_2_5_vl video processing #6868

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

Merged
merged 4 commits into from
Feb 11, 2025

Conversation

JJJYmmm
Copy link
Contributor

@JJJYmmm JJJYmmm commented Feb 8, 2025

What does this PR do?

Fixes #6860

Calculate the actual fps for each video and get correct second_per_grid_ts

The origin solution directly use video_fps, this may be incorrect when meeting long video.

The modified version works on my project.

Before submitting

@JosonChan1998
Copy link

@JJJYmmm Hi, have you tested the effect of this implementation? In SFT stage?

@JJJYmmm
Copy link
Contributor Author

JJJYmmm commented Feb 10, 2025

@JJJYmmm Hi, have you tested the effect of this implementation? In SFT stage?

Not yet, I just refer the solution of official repo.

@hiyouga hiyouga self-requested a review February 11, 2025 05:56
Copy link
Owner

@hiyouga hiyouga left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@hiyouga hiyouga merged commit 9153a7b into hiyouga:main Feb 11, 2025
12 checks passed
@hiyouga hiyouga added the solved This problem has been already solved label Feb 11, 2025
@JJJYmmm JJJYmmm deleted the fix-qwen_2_5_vl-video-processing branch February 11, 2025 08:18
1587causalai pushed a commit to 1587causalai/llama_factory that referenced this pull request Feb 18, 2025
* fix qwen_2_5_vl video processing

* Update mm_plugin.py

* Update mm_plugin.py

---------

Co-authored-by: hoshi-hiyouga <[email protected]>
stephen-nju pushed a commit to stephen-nju/Llmtrain that referenced this pull request Mar 24, 2025
* fix qwen_2_5_vl video processing

* Update mm_plugin.py

* Update mm_plugin.py

---------

Co-authored-by: hoshi-hiyouga <[email protected]>
Former-commit-id: 9153a7b
yoonseok312 pushed a commit to pensieve-ai/LLaMA-Factory-vlm that referenced this pull request Apr 29, 2025
* fix qwen_2_5_vl video processing

* Update mm_plugin.py

* Update mm_plugin.py

---------

Co-authored-by: hoshi-hiyouga <[email protected]>
Former-commit-id: 9153a7b
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
solved This problem has been already solved
Projects
None yet
Development

Successfully merging this pull request may close these issues.

qwen2.5_vl vision info process
3 participants