MCRGNet: Morphological Classification of Radio Galaxy Network

Discussion topics for individual codes
Post Reply
Ada Coda
ASCL Robot
Posts: 1466
Joined: Thu May 08, 2014 5:37 am

MCRGNet: Morphological Classification of Radio Galaxy Network

Post by Ada Coda » Sat Jul 27, 2019 7:19 pm

MCRGNet: Morphological Classification of Radio Galaxy Network

Abstract: MCRGNet (Morphological Classification of Radio Galaxy Network) classifies radio galaxies of different morphologies. It is based on the Convolutional Neural Network (CNN), which is trained and applied under a three-step framework: 1.) pretraining the network unsupervisedly with unlabeled samples, 2.) fine-tuning the pretrained network parameters supervisedly with labeled samples, and 3.) classifying a new radio galaxy by the trained network. The code uses a dichotomous tree classifier composed of cascaded CNN based subclassifiers.

Credit: Ma, Zhixian

Site: https://github.com/myinxd/mcrgnet
https://ui.adsabs.harvard.edu/abs/2019ApJS..240...34M

Bibcode: 2019ascl.soft07026M

ID: ascl:1907.026

Post Reply