Skip to content

Latest commit

 

History

History
11 lines (9 loc) · 739 Bytes

README.md

File metadata and controls

11 lines (9 loc) · 739 Bytes

Normalizing flows with PyTorch

Implementation and tutorials of normalizing flows with the novel distributions module. The current set of tutorials and implementations is

  1. Implementing and optimizing planar flows
  2. Different types of invertible flows (radial, batchnorm, affine)
  3. Using flows in variational inference (VAEs)
  4. Auto-regressive types of flows (RealNVP, MAF, IAF) Still very much drafty work in progress
  5. More advanced and recent type of flows

Sorry everyone for the very long delay, we shall try to finish this tutorial session with new advances in generative flows (GLOW) and more advanced ideas (NODE, FFJORD) in the upcoming weeks :)