Skip to main content

Neural Network-Boosted Importance Sampling for Bayesian Statistics

Project description

Logo

Unit Testing Status Documentation Status Code Coverage PyPI PyPI - Downloads Conda Conda - Downloads License: MIT Language: Python

Nautilus is an MIT-licensed pure-Python package for Bayesian posterior and evidence estimation. It utilizes importance sampling and efficient space exploration using neural networks. Compared to traditional MCMC and Nested Sampling codes, it often needs fewer likelihood calls and produces much larger posterior samples. Additionally, nautilus is highly accurate and produces Bayesian evidence estimates with percent precision. It is widely used in many areas of astrophysical research.

Example

This example, sampling a 3-dimensional Gaussian, illustrates how to use nautilus.

import corner
import numpy as np
from nautilus import Prior, Sampler
from scipy.stats import multivariate_normal

prior = Prior()
for key in 'abc':
    prior.add_parameter(key)

def likelihood(param_dict):
    x = [param_dict[key] for key in 'abc']
    return multivariate_normal.logpdf(x, mean=[0.4, 0.5, 0.6], cov=0.01)

sampler = Sampler(prior, likelihood)
sampler.run(verbose=True)
points, log_w, log_l = sampler.posterior()
corner.corner(points, weights=np.exp(log_w), labels='abc')

Installation

The most recent stable version of nautilus is listed in the Python Package Index (PyPI) and can be installed via pip.

pip install nautilus-sampler

Additionally, nautilus is also on conda-forge. To install via conda use the following command.

conda install -c conda-forge nautilus-sampler

Documentation

You can find the documentation at nautilus-sampler.readthedocs.io.

Attribution

A paper describing nautilus's underlying methods and performance has been published in the Monthly Notices of the Royal Astronomical Society. A draft of the paper is also available on arXiv. Please cite the paper if you find nautilus helpful in your research.

@article{nautilus,
    author = {Lange, Johannes U},
    title = "{nautilus: boosting Bayesian importance nested sampling with deep learning}",
    journal = {Monthly Notices of the Royal Astronomical Society},
    volume = {525},
    number = {2},
    pages = {3181-3194},
    year = {2023},
    month = {08},
    doi = {10.1093/mnras/stad2441},
    url = {https://doi.org/10.1093/mnras/stad2441},
    eprint = {https://academic.oup.com/mnras/article-pdf/525/2/3181/51331635/stad2441.pdf},
}

License

Nautilus is licensed under the MIT License. The logo uses an image from the Illustris Collaboration.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nautilus_sampler-1.0.5.tar.gz (41.5 kB view details)

Uploaded Source

Built Distribution

nautilus_sampler-1.0.5-py3-none-any.whl (33.1 kB view details)

Uploaded Python 3

File details

Details for the file nautilus_sampler-1.0.5.tar.gz.

File metadata

  • Download URL: nautilus_sampler-1.0.5.tar.gz
  • Upload date:
  • Size: 41.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: python-requests/2.32.3

File hashes

Hashes for nautilus_sampler-1.0.5.tar.gz
Algorithm Hash digest
SHA256 1247eb633ac18fddafb85f37015ba0b3b5de3a4afe12ad09bfdf95755ecfe3ac
MD5 5461aa231e54931186181f3ff52dbf29
BLAKE2b-256 8e3193eae1e2c19949f1a69c0f58ef460b8e8c568aedd74b3a848ed927e4a5ba

See more details on using hashes here.

File details

Details for the file nautilus_sampler-1.0.5-py3-none-any.whl.

File metadata

File hashes

Hashes for nautilus_sampler-1.0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 4351986f938ba897dc850a9cac2894509df6d4d0a19dc3edd0070032ef0b016e
MD5 6e50496089d2a35cfbb592774e161aaf
BLAKE2b-256 00951ff5b8b04b1fde075771023786a56878599563ad0f1e3aac99215420ffb1

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page