celerite2

celerite is an algorithm for fast and scalable Gaussian Process (GP) Regression in one dimension and this library, celerite2 is a re-write of the original celerite project to improve numerical stability and integration with various machine learning frameworks. This implementation includes interfaces in Python and C++, with full support for PyMC (v3 and v4) and JAX.

This documentation won’t teach you the fundamentals of GP modeling but the best resource for learning about this is available for free online: Rasmussen & Williams (2006). Similarly, the celerite algorithm is restricted to a specific class of covariance functions (see the original paper for more information and a recent generalization for extensions to structured two-dimensional data). If you need scalable GPs with more general covariance functions, GPyTorch or tinygp might be good choices.

celerite2 is being actively developed in a public repository on GitHub so if you have any trouble, open an issue there.

User Guide

• Installation
  • Using pip
  • From Source
  • Testing
• Upgrading from celerite
  • Why does this even exist?
  • What interfaces are available in celerite2?
  • A simple example
• Citing celerite2

Tutorials

• Getting started
  • Maximum likelihood
  • Posterior inference using emcee
  • Posterior inference using PyMC
  • Posterior inference using numpyro
  • Comparison

API Details

• Python interface
  • GaussianProcess
  • Model building
• PyMC3 interface
  • GaussianProcess
  • PyMC3 support
• PyMC4 interface
  • GaussianProcess
  • PyMC (v4) support
• JAX interface
  • GaussianProcess
  • numpyro support
• C++ interface
  • Basic interface
  • Autodiff Interface
  • Terms

License & attribution

Copyright 2020-2022 Daniel Foreman-Mackey.

The source code is made available under the terms of the MIT license.

If you make use of this code, please cite the relevant papers: Citing celerite2

Changelog