open_clip icon indicating copy to clipboard operation
open_clip copied to clipboard
verified publisher icon mlfoundations

How to initialize the encoder to pretrain CLIP？

Open shihuai opened this issue 1 year ago • 0 comments

HI, Thank you for your open source work. I am also very interested in your work. Regarding your work, I have a question: When you use Covnext as the image encoder to pre-train CLIP, the weights of convnext use ImageNet pre-trained initialization or random initialization?

shihuai avatar Apr 12 '24 04:04 shihuai