LAION-AI / LAION-AI/CLIP_benchmark
Support computing metrics using test time distribution normalisation
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 814
- Forks
- 103
- PR merge metrics
- No merged PRs in 30d
Description
https://twitter.com/YifeiZhou02/status/1716513495087472880?t=tktnVDniUhDadYYwEM3F5w&s=19 claims by using a second term in the distance computation they improve image net zero shot classification accuracy by like 2pp on imagenet for B/32
I'm curious if that's true as well for better models
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the linked X/Twitter post to identify the proposed second term and test-time distribution normalization. Then locate the benchmark's existing zero-shot ImageNet evaluation entry point and determine how the method can be compared with the current metrics; done means reporting whether the claimed accuracy improvement holds for better models.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100