The tiniest of Gaussian Process libraries.

tinygp is an extremely lightweight library for Gaussian Process (GP) regression in Python, built on top of jax. It is not (yet?) designed to provide all the shiniest algorithms for scalable computations (check out celerite2 or GPyTorch if you need something like that), but I think it has a nice interface and itโ€™s pretty fast. Thanks to jax, tinygp supports things like GPU acceleration and automatic differentiation.

How to find your way around?

๐Ÿ‘‰ A good place to get started is with the installation guide and then the tutorials.

๐Ÿ‘‰ For all the details, check out the full API documentation, including a list of all the built-in kernel functions.

๐Ÿ‘‰ If you find bugs or otherwise have trouble getting tinygp to do what you want, head on over to the GitHub issues page.

Authors & Licenseยถ

Copyright 2021 Dan Foreman-Mackey

Built by Dan Foreman-Mackey and contributors (see the contribution graph for the most up to date list). Licensed under the MIT license (see LICENSE).