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

Add support for evals using concatenation of CLS + mean patch tokens as input features #742

Open
nkaenzig opened this issue Jan 27, 2025 · 0 comments · May be fixed by #743
Open

Add support for evals using concatenation of CLS + mean patch tokens as input features #742

nkaenzig opened this issue Jan 27, 2025 · 0 comments · May be fixed by #743
Assignees
@nkaenzig

Comments

@nkaenzig
Copy link
Collaborator

nkaenzig commented Jan 27, 2025 •
edited
Loading

Instead of using the only CLS token embedding as input features, a common approach in literature is to concatenate the CLS embedding with the mean aggregation of all patch token embeddings.
@nkaenzig nkaenzig self-assigned this Jan 27, 2025
@nkaenzig nkaenzig linked a pull request Jan 27, 2025 that will close this issue
Draft
@nkaenzig nkaenzig linked a pull request Jan 27, 2025 that will close this issue
Add concat_mean_patch_tokens option to models in registry #743
Draft
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@nkaenzig nkaenzig

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Add concat_mean_patch_tokens option to models in registry kaiko-ai/eva
1 participant
@nkaenzig