https://github.com/karpathy/pytorch-normalizing-flows

Normalizing flows in PyTorch. Current intended use is education not production.
https://github.com/karpathy/pytorch-normalizing-flows

Last synced: 11 days ago
JSON representation

Normalizing flows in PyTorch. Current intended use is education not production.

• Host: GitHub
• URL: https://github.com/karpathy/pytorch-normalizing-flows
• Owner: karpathy
• Created: 2019-12-09T17:55:39.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2020-01-27T18:37:10.000Z (almost 5 years ago)
• Last Synced: 2024-07-31T18:15:58.383Z (3 months ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 2.11 MB
• Stars: 834
• Watchers: 27
• Forks: 101
• Open Issues: 9
• Metadata Files:
  • Readme: Readme.md

Awesome Lists containing this project

• awesome-energy-based-model - `pytorch-normalizing-flows`
• awesome-normalizing-flows - pytorch-normalizing-flows

README

        
# pytorch-normalizing-flows

Implementions of normalizing flows (NICE, RealNVP, MAF, IAF, Neural Splines Flows, etc) in PyTorch.

![Normalizing Flow fitting a 2D dataset](https://github.com/karpathy/pytorch-normalizing-flows/blob/master/assets/moon_flow.png)

**todos**

- TODO: make work on GPU

- TODO: 2D -> ND: get (flat) using MNIST

- TODO: ND -> images (multi-scale architectures, Glow nets, etc) on MNIST/CIFAR/ImageNet

- TODO: more stable residual-like IAF-style updates (tried but didn't work too well)

- TODO: parallel wavenet

- TODO: radial/planar 2D flows from Rezende Mohamed 2015?