yang-song/score_sde_pytorch

Question about conditional generation

Icamd opened this issue · 0 comments

Thank you for your paper and code! I am trying to perform conditional generation using score-based diffusion model. Is it possible to apply FiLM (Feature-wise Linear Modulation) conditioning [Film: Visual reasoning with a general conditioning layer] on score-based diffusion model? And how could I edit the normal DDPM models into score-based generative models? Thank you!