improved-diffusion
improved-diffusion copied to clipboard
openai
Release for Improved Denoising Diffusion Probabilistic Models
Results
112
improved-diffusion issues
Sort by
recently updated
recently updated
newest added
In my opinion, authors define L_{vlb} = L_0 + ... + L_T, not L_t. Thus, they may calculate the vlb loss with scale factor T (self.num_timestep). _Originally posted by @yhy258...
If I fine-tune the pre-trained [256x256_diffusion_uncond.pt] model on a small dataset (e.g., only 5K images), what key considerations should I keep in mind?
