Scalable inference for a generative model of astronomical images

First Commit


Last Touched

2 days ago

Commit Count

2233 commits



Build Status Coverage Status

Celeste.jl finds and characterizes stars and galaxies in astronomical images. It implements variational inference for the generative model described in

Jeffrey Regier, Andrew Miller, Jon McAuliffe, Ryan Adams, Matt Hoffman, Dustin Lang, David Schlegel, and Prabhat. “Celeste: Variational inference for a generative model of astronomical images”. In: Proceedings of the 32nd International Conference on Machine Learning (ICML). 2015.


The main entry point is bin/celeste.jl. Run celeste.jl --help for detailed usage information.

Note that in the score mode, the script requires data downloaded from the CasJobs Stripe82 database in a given RA, Dec range. See here for information on downloading this data from the SDSS CasJobs server.


Celeste.jl is free software, licensed under version 2.0 of the Apache License.