Normalizing Flows are generative models which produce tractable distributions where both sampling and density evaluation can be efficient and exact. This repository contains examples of implementation of NF.
Many thanks to the wonderful 'Probabilistic Deep Learning' course provided by Dr.Kevin Webster (Imperial College London), from where I get started in learning this topics.
Description:
The NF-distribution-2d notebook file implements a NF to estimate a 2-dimensional distribution. The specific NF used here is called the Masked Autoregressive Flow.