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

Bad test results #23

Open
liziming5353 opened this issue Nov 13, 2023 · 6 comments
Open

Bad test results #23

liziming5353 opened this issue Nov 13, 2023 · 6 comments

Comments

@liziming5353
Copy link

I download llama1 and use apply_delta.py to generate ckpt. Then I use the original run_valley.py without any modifications to test on a test dataset, but get terrible results. May I ask if there is anything special that needs to be set up? Like prompt or something?

@RupertLuo
Copy link
Owner

RupertLuo commented Nov 13, 2023 via email

@RupertLuo
Copy link
Owner

RupertLuo commented Nov 13, 2023 via email

@liziming5353
Copy link
Author

The query is:
image
The response is:
image
I also print the outputs before self.process_response, and the result is:
image
No matter which video I input, it's always this result.

@RupertLuo
Copy link
Owner

Are your transformers version 4.28.0.dev?

@pratikkayal
Copy link

pratikkayal commented Mar 12, 2024

Hi I'm getting the same issue, I do have transformers version 4.28.0.dev Getting the following output: [',,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,....,,,,,,,,,,,,,,,,,,,,,,,,....................................................................................................,.,.........,.,,,,,,.,................................................................,,,,,,,.,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,................................................................................,.,.,...............................................................................................................................................................................,.,.,.,.....,.,.,.,.,................................................................................................................................................................................................,.,....,.,.,.,............................................................ (G..................................................................................-------']

@annopackage
Copy link

same issue here, offline demo seems normal.

The query is: image The response is: image I also print the outputs before self.process_response, and the result is: image No matter which video I input, it's always this result.

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

4 participants