Public API#

The following pages describe the technical details of all the public-facing members of the tinygp API. This isn’t meant to be introductory and, if you’re new here, the Tutorials might be a better place to start. That being said, we’ve tried to provide sufficiently detailed descriptions of all the provided methods for once you (/we) get into the weeds. Please open issues or pull requests if you find anything lacking.

Primary Interface#

tinygp is an extremely lightweight library for building Gaussian Process models in Python, built on top of jax. The primary way that you will use to interact with tinygp is by constructing “kernel” functions using the building blocks provided in the kernels subpackage (see kernels package), and then passing that to a GaussianProcess object to do all the computations. Check out the Tutorials for a more complete introduction.

GaussianProcess(kernel, X, *[, diag, noise, ...])

An interface for designing a Gaussian Process regression model

gp.ConditionResult(log_probability, gp)

The result of conditioning a GaussianProcess on data