olllom/nflows

Normalizing flows in PyTorch

nflows

nflows is a comprehensive collection of normalizing flows using PyTorch.

Installation

To install from PyPI:

pip install nflows

Usage

To define a flow:

from nflows import transforms, distributions, flows

# Define an invertible transformation.
transform = transforms.CompositeTranform([
    transforms.AffineCouplingTransform(...),
    transforms.RandomPermutation(...)
])

# Define a base distribution.
base_distribution = distributions.StandardNormal(...)

# Combine into a flow.
flow = flows.Flow(transform=transform, distribution=base_distribution)

To evaluate log probabilities of inputs:

log_prob = flow.log_prob(inputs)

To sample from the flow:

samples = flow.sample(num_samples)

Additional examples of the workflow are provided in examples folder.

Development

You can install all the dependencies using the environment.yml file to create a conda environment:

conda env create -f environment.yml

Alternatively, you can install via setup.py (the dev flag installs development and testing dependencies):

pip install -e ".[dev]"

References

nflows is derived from bayesiains/nsf originally published with

C. Durkan, A. Bekasov, I. Murray, G. Papamakarios, Neural Spline Flows, NeurIPS 2019. [arXiv] [bibtex]

nflows has been used in

Conor Durkan, Iain Murray, George Papamakarios, On Contrastive Learning for Likelihood-free Inference, ICML 2020. [arXiv].

Artur Bekasov, Iain Murray, Ordering Dimensions with Nested Dropout Normalizing Flows. [arXiv].

Tim Dockhorn, James A. Ritchie, Yaoliang Yu, Iain Murray, Density Deconvolution with Normalizing Flows. [arXiv].

nflows is used by the conditional density estimation package pyknos, and in turn the likelihood-free inference framework sbi.