How to get all the features of the image encoder and text encoder

openai / CLIP

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

MIT License

24.55k stars 3.2k forks source link

Open wuyukun-tong opened 5 months ago

wuyukun-tong commented 5 months ago

I want to extract more features than just the 512 dimensional cls token from the CLIP pre-trained model, how can I modify that？