AstroPhot

<picture> <source media="(prefers-color-scheme: dark)" srcset="https://github.com/Autostronomy/AstroPhot/blob/main/media/AP_logo_white.png?raw=true"> <source media="(prefers-color-scheme: light)" srcset="https://github.com/Autostronomy/AstroPhot/blob/main/media/AP_logo.png?raw=true"> <img alt="AstroPhot logo" src="media/AP_logo.png" width="70%"> </picture>

AstroPhot is a fast, flexible, and principled astronomical image modelling tool
for precise parallel multi-wavelength/epoch photometry. It is a python based
package that uses PyTorch or JAX to quickly and efficiently perform analysis
tasks. Written by Connor Stone for tasks such as
LSB imaging, handling crowded fields, multi-band photometry, and analyzing
massive data from future telescopes. AstroPhot is flexible and fast for any
parametric astronomical image modelling task. While it uses PyTorch and/or JAX
(originally developed for Machine Learning) it is NOT a machine learning based
tool. In fact AstroPhot very rigidly sticks to Gaussian/Poisson likelihood
modelling (with extensions for priors if desired).

AstroPhot is now a caskade ecosystem project,
meaning its parameters have an incredible amount of flexibility. Check out the
documentation for more details!

Installation

AstroPhot can be installed with pip:

pip install astrophot

If PyTorch gives you any trouble on your system, just follow the instructions on
the pytorch website to install a version for your
system.

Also note that AstroPhot is only available for python3.

See the documentation for more details.

Documentation

You can find the documentation at the
ReadTheDocs site connected with the AstroPhot project
which covers many of the main use cases for AstroPhot. There is tons of useful
information in there, hopefully you can mix and match tutorials to get to just
about any parametric image modelling task quickly! Feel free to contact the
author, Connor Stone, for any questions not
answered by the documentation or tutorials.

Credit / Citation

If you use AstroPhot in your research, please follow the
citation instructions here.

Thanks to our contributors!