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.

Sign up for GitHub

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

Jump to bottom

Setting for Zero-shot ImageNet classification #53

Closed
zfchenUnique opened this issue Jun 16, 2023 · 3 comments
Closed

Setting for Zero-shot ImageNet classification #53

zfchenUnique opened this issue Jun 16, 2023 · 3 comments

Comments

@zfchenUnique
Copy link

Hi, thanks for your great work!

Could you please provide your evaluation code for zero-shot imageNet classification?
More specifically, I would like to know how to implement "limit the candidate tokens during each token prediction, such that the predicted result is guaranteed to be within the vocabulary."
Thanks in advance.

Sincerely,
Zhenfang
@amsword
Copy link
Contributor

amsword commented Jun 30, 2023

That is based on a trie structure. I will try to find out some time to clean those codes.

@vishaal27
Copy link

Any update on this?

@amsword
Copy link
Contributor

amsword commented Sep 28, 2023

please uncomment the following to enable this decoding

GenerativeImage2Text/generativeimage2text/model.py

Line 42 in faae4fb

#from .trie_decoder import TrieAutoRegressiveBeamSearch, get_trie

@amsword amsword closed this as completed Sep 28, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@amsword @zfchenUnique @vishaal27