encord-team/text-to-image-eval

Evaluate custom and HuggingFace text-to-image/zero-shot-image-classification models like CLIP, SigLIP, DFN5B, and EVA-CLIP. Metrics include Zero-shot accuracy, Linear Probe, Image retrieval, and KNN accuracy.

1 best points · 1 comments · 1 Hacker News post · 58 GitHub stars · View on GitHub

Hacker News discussions

• We Built an Open-Source Text-to-Image Evaluation Library for Clip Models (1 points, 1 comments, by Stephen_Oladele, 2024-05-07)