dx-dtran / gpt2-vision

Convert GPT-2 into a multimodal model using CLIP. Under 1000 lines of pure PyTorch

Geek Repo

Github PK Tool