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

Few shot training with SAA+ #33

Open
zeddo123 opened this issue Mar 20, 2024 · 1 comment
Open

Few shot training with SAA+ #33

zeddo123 opened this issue Mar 20, 2024 · 1 comment

Comments

@zeddo123
Copy link

zeddo123 commented Mar 20, 2024

Hello,

I've been reading the paper on the model and there isn't much mention of Few Shot Learning in it. However, in the code there's an argument k_shot, is this referring to few shot learning? if so could you shed some light on how it is implemented in this context? If not how it might be done it SAA+?

@caoyunkang
Copy link
Owner

Thank you for your question! Currently, SAA+ does not support few-shot fine-tuning. Fine-tuning is highly dependent on Grounding-DINO, which has not yet released its training/fine-tuning code.

In the meantime, we encourage you to explore our latest work, AdaCLIP, where you can find the training process available in the repository.

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