https://github.com/Karami-m/Invertible-Convolutional-Flow

Source code for the paper: M. Karami, D. Schuurmans, Jascha Sohl-Dickstein, Daniel Duckworth, Laurent Dinh, “Invertible Convolutional Flow", Advances in Neural Information Processing Systems (NeurIPS) 2019
https://github.com/Karami-m/Invertible-Convolutional-Flow

Last synced: 3 months ago
JSON representation

Source code for the paper: M. Karami, D. Schuurmans, Jascha Sohl-Dickstein, Daniel Duckworth, Laurent Dinh, “Invertible Convolutional Flow", Advances in Neural Information Processing Systems (NeurIPS) 2019

• Host: GitHub
• URL: https://github.com/Karami-m/Invertible-Convolutional-Flow
• Owner: Karami-m
• License: mit
• Created: 2019-10-28T09:33:48.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2019-10-28T10:38:12.000Z (about 5 years ago)
• Last Synced: 2024-07-12T21:15:48.988Z (4 months ago)
• Homepage:
• Size: 10.7 KB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-normalizing-flows - [Code

README

        
# Invertible-Convolutional-Flow

Source code for reproducing the results in the paper: 

>M. Karami, D. Schuurmans, Jascha Sohl-Dickstein, Daniel Duckworth, Laurent Dinh, “Invertible Convolutional Flow", Advances in Neural Information Processing Systems (NeurIPS) 2019

> [[arxiv]](https://arxiv.org/.../...) [[bibtex]](http://...)

## Requirements

 - Tensorflow (tested with v1.8.0)

 - Pytorch (tested with ...)

## Usage

Tabular datasets from [MAF](https://github.com/gpapamak/maf):

...

MNIST/CIFAR10:

...

VAE Experiments (based on [Sylvester VAE](https://github.com/riannevdberg/sylvester-flows)):

...

## Download datasets

### Tabular (UCI + BSDS300)

...

### VAE datasets

...