Skip to content

README.md

Latest commit

 

History

History
20 lines (15 loc) · 576 Bytes

README.md

File metadata and controls

20 lines (15 loc) · 576 Bytes

pure torch-based Diffusion

The Diffusion model, which can be run by simply installing torch, helps us learn the main components of the algorithm. The code is extremely simple and readable, you can simply read the code and comments to learn everything.

TODO

  • Add VAE to the diffusion model.
  • Implement parameter configuration.
  • Implement diffusion with DiT.
  • Implement VAR model.

DDPM Train

python ddpm_train.py

DDPM Inference

python ddpm_inference.py

VQVAE Train

python vqvae_train.py

VQVAE Inference

python vqvae_inference.py