gpjax 0.11.2

Gaussian processes in JAX.

Quickstart | Install guide | Documentation | Slack Community

GPJax aims to provide a low-level interface to Gaussian process (GP) models in Jax, structured to give researchers maximum flexibility in extending the code to suit their own needs. The idea is that the code should be as close as possible to the maths we write on paper when working with GP models.

Package organisation

Contributions

We would be delighted to receive contributions from interested individuals and groups. To learn how you can get involved, please read our guide for contributing. If you have any questions, we encourage you to open an issue. For broader conversations, such as best GP fitting practices or questions about the mathematics of GPs, we invite you to open a discussion.

Another way you can contribute to GPJax is through issue triaging. This can include reproducing bug reports, asking for vital information such as version numbers and reproduction instructions, or identifying stale issues. If you would like to begin triaging issues, an easy way to get started is to subscribe to GPJax on CodeTriage.

As a contributor to GPJax, you are expected to abide by our code of conduct. If you feel that you have either experienced or witnessed behaviour that violates this standard, then we ask that you report any such behaviours through this form or reach out to one of the project's gardeners.

Feel free to join our Slack Channel, where we can discuss the development of GPJax and broader support for Gaussian process modelling.

Governance

GPJax was founded by Thomas Pinder. Today, the project's gardeners are daniel-dodd@, henrymoss@, st--@, and thomaspinder@, listed in alphabetical order. The full governance structure of GPJax is detailed here. We appreciate all the contributors to GPJax who have helped to shape GPJax into the package it is today.

Supported methods and interfaces

Notebook examples

• Conjugate Inference
• Classification
• Sparse Variational Inference
• Stochastic Variational Inference
• Laplace Approximation
• Inference on Non-Euclidean Spaces
• Inference on Graphs
• Pathwise Sampling
• Learning Gaussian Process Barycentres
• Deep Kernel Regression
• Poisson Regression
• Bayesian Optimisation

Guides for customisation

• Custom kernels
• UCI regression

Conversion between .ipynb and .py

Above examples are stored in examples directory in the double percent (py:percent) format. Checkout jupytext using-cli for more info.

• To convert example.py to example.ipynb, run:

jupytext --to notebook example.py

• To convert example.ipynb to example.py, run:

jupytext --to py:percent example.ipynb

Installation

Stable version

The latest stable version of GPJax can be installed via pip:

pip install gpjax

Note

We recommend you check your installation version:

python -c 'import gpjax; print(gpjax.__version__)'

Development version

Warning

This version is possibly unstable and may contain bugs.

Note

We advise you create virtual environment before installing:

conda create -n gpjax_experimental python=3.10.0
conda activate gpjax_experimental

Clone a copy of the repository to your local machine and run the setup configuration in development mode.

git clone https://github.com/JaxGaussianProcesses/GPJax.git
cd GPJax
hatch env create
hatch shell

We recommend you check your installation passes the supplied unit tests:

hatch run dev:test

Citing GPJax

If you use GPJax in your research, please cite our JOSS paper.

@article{Pinder2022,
  doi = {10.21105/joss.04455},
  url = {https://doi.org/10.21105/joss.04455},
  year = {2022},
  publisher = {The Open Journal},
  volume = {7},
  number = {75},
  pages = {4455},
  author = {Thomas Pinder and Daniel Dodd},
  title = {GPJax: A Gaussian Process Framework in JAX},
  journal = {Journal of Open Source Software}
}