# pollux

## Introduction

Pollux is a Python library for constructing generative models of astronomical spectra
and other kinds of data. It is built on top of [JAX][jax] and is designed for use in
probabilistic and machine learning contexts.

Two classes of models are currently supported:

- [_Lux_](https://arxiv.org/abs/2502.01745): Multi-output, generative, latent variable
  models for inferring embedded representations of spectroscopic and many other kinds of
  data.
- [_Cannon_](https://arxiv.org/abs/1501.07604): Data-driven models for inferring stellar
  parameters, element abundances, and other labels from stellar spectra.

---

## Installation

Install `pollux` with `pip` directly from the GitHub repository:

```bash
pip install git+https://github.com/adrn/pollux
```

In the future, we plan to provide stable releases through PyPI.

<!-- [![PyPI version][pypi-version]][pypi-link] -->
<!-- [![PyPI platforms][pypi-platforms]][pypi-link] -->

[jax]: https://jax.readthedocs.io/en/latest/

<!-- [pypi-link]: https://pypi.org/project/TODO/ -->
<!-- [pypi-platforms]: https://img.shields.io/pypi/pyversions/TODO -->
<!-- [pypi-version]: https://img.shields.io/pypi/v/TODO -->

## Get Started

The best way to get started with `pollux` is to work through the tutorials:

```{eval-rst}
.. include:: _tutorials.rst
```

```{toctree}
:maxdepth: 1
:hidden:
:caption: API Reference

api/index.md
```