OpenGVLab / OpenGVLab/Ask-Anything

Problems when running /video_chat/demo.ipynb

Open
#261 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
3.4k
Forks
268
PR merge metrics
No merged PRs in 30d

Description

I try to run ./video_chat/demo.ipynb with config_7b.json according to README.md. There are massive incompatibleKeys warnings during load_state_dict of eva_vit_g/blip2_pretrained_flant5xxl and videochat_7b.pth. The answer is also in chaos. I try to run pure text QA without uploading videos and it runs well so the LLM part may be OK.
I don't know which step I didn't do right. Here is the download url i used. Really appreciate your help!

eva_vit_g: https://storage.googleapis.com/sfr-vision-language-research/LAVIS/models/BLIP2/eva_vit_g.pth`
blip2_pretrained_flant5xxl: https://storage.googleapis.com/sfr-vision-language-research/LAVIS/models/BLIP2/blip2_pretrained_flant5xxl.pth`
videochat_7b: https://drive.google.com/file/d/1C4s65TC5Zr85I8dZmnfrrw6oDAjj1H4P/view?usp=sharing

Image

Image

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with README.md and reproduce ./video_chat/demo.ipynb using config_7b.json, recording the load_state_dict warnings for eva_vit_g, blip2_pretrained_flant5xxl, and videochat_7b.pth. Compare the video run with the reported working pure-text QA path; done means the notebook loads the supplied models without the incompatible-key problem and produces coherent video answers.

Written by the indexing model from the issue text.

Assessment

Tech stack
jupyter-notebook, python
Domain
computer-vision, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.