Diffusion

Diffusion copied to clipboard

Use vae to encode x and then train vae = AutoencoderKL.from_pretrained(f"stabilityai/sd-vae-ft-ema").to(device) x = vae.encode(x).latent_dist.sample().mul_(0.18215) Sample with torch.no_grad(): z = ema_sample_method(opt.n_sample, z_shape, guide_w=opt.w) x_gen = vae.decode(z / 0.18215).sample The generation effect...

yh-xxx

A request for the pretrained model weights.

2

Thank you for your excellent work！ May I ask if it's possible to release the trained models? (Class-conditional DDPM & DDIM)

zhaisf