Trainable pytorch implementations of CLIP, DDPM, VQGAN and Stable Diffusion from scratch. Uses CIFAR-10 64x64 dataset, more dataset support to be added later. Current results can be viewed in 'results' directory
songyang86 / stable-diffusion-pytorch-training Goto Github PK
View Code? Open in Web Editor NEWThis project forked from dhillon24/stable-diffusion-pytorch-training
Trainable PyTorch implementation of CLIP, DDPM, VQGAN and Stable Diffusion