[GH-ISSUE #502] Recall (R@k) way lower than the one obtained in papers #426

Open
opened 2026-10-31 21:14:52 +00:00 by bsakaguchi · 1 comment
Owner

Originally created by @asczdvefb42 on GitHub (Jun 11, 2025).
Original GitHub issue: https://github.com/openai/CLIP/issues/502

Here is what i do :

  • Get test dataset of image - caption (IIITD-20K dataset)
  • calculate embeddings with my fine tuned CLIP
  • calculate cosine distance between each text to all images
  • get the k closest images to the text, if the corresponding image is in it, do +1 to score
  • get the recall by dividing the score by the length of the test dataset.

This is my recall at k. I obtain a R@1 of 17%, while most papers when finetuning CLIP obtain at least 60% recall at 1. Any idea what i could be doing wrong?

Originally created by @asczdvefb42 on GitHub (Jun 11, 2025). Original GitHub issue: https://github.com/openai/CLIP/issues/502 Here is what i do : - Get test dataset of image - caption (IIITD-20K dataset) - calculate embeddings with my fine tuned CLIP - calculate cosine distance between each text to all images - get the k closest images to the text, if the corresponding image is in it, do +1 to score - get the recall by dividing the score by the length of the test dataset. This is my recall at k. I obtain a R@1 of 17%, while most papers when finetuning CLIP obtain at least 60% recall at 1. Any idea what i could be doing wrong?
Author
Owner

@ZJ331456 commented on GitHub (Aug 23, 2025):

Any fix for this? I've been running into the same problem recently.

<!-- gh-comment-id:3217236919 --> @ZJ331456 commented on GitHub (Aug 23, 2025): Any fix for this? I've been running into the same problem recently.
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
openai/CLIP#426
No description provided.