Astrophysics Source Code Library

Making codes discoverable since 1999

ASCL Code Record

[ascl:2106.012] StarcNet: Convolutional neural network for classifying galaxy images into morphological classes

StarcNet (STAR Cluster classification NETwork) classifies star clusters from galaxy images taken by the Hubble Space Telescope (HST); it uses a convolutional neural network (CNN) trained to classify five-band galaxy images into four morphological classes. Written in PyTorch, StarcNet runs using mosaics (.fits files with the galaxy photometric information) and catalogs (.tab files with object coordinates), and includes the option to also download the galaxy mosaics from a single .tar.gz file per galaxy (as from the Legacy ExtraGalactic UV Survey).

Code site:
Described in:

Views: 360

Add this shield to your page
Copy the above HTML to add this shield to your code's website.