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

Can we test without loading pre-trained weights? #21

Open
liyiyue2021 opened this issue Jan 4, 2024 · 4 comments
Open

Can we test without loading pre-trained weights? #21

liyiyue2021 opened this issue Jan 4, 2024 · 4 comments

Comments

@liyiyue2021
Copy link

Can I test without loading pre-trained weights, and how do I set the checkpoint_path?

@ByChelsea
Copy link
Owner

You must have pre-trained weights to perform the test. You can set either an absolute or relative path, just by changing the checkpoint_path in test_zero_shot.sh.

@liyiyue2021
Copy link
Author

Thank you for your very prompt reply. What I mean is, is it possible to train the model without loading weights from other pre-trained data (eg, Visa or MVTec datasets)? I would like to test it using the original CLIP weights.

@ByChelsea
Copy link
Owner

As mentioned in WinCLIP, directly using the weights of CLIP does not yield good results (the first line of Table 8). Are you referring to this experiment?

@liyiyue2021
Copy link
Author

Got it, thank you.

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

2 participants