ASCL.net

Astrophysics Source Code Library

Making codes discoverable since 1999

ASCL Code Record

[ascl:2112.019] O'TRAIN: Optical TRAnsient Identification NEtwork

The O'TRAIN package identifies transients in astronomical images based on a Convolutional Neural Network (CNN). It works on images from different telescopes and, through the use of Docker, can be deployed on different operating systems. O'TRAIN uses image cutouts containing real and false transients provided by the user to train a CNN algorithm implemented with Keras. Built-in diagnostics help to characterize the accuracy of the training, and a trained model is used to classify any new cutouts.

Code site:
https://github.com/dcorre/otrain
Described in:
https://ui.adsabs.harvard.edu/abs/2021arXiv211210280M
Bibcode:
2021ascl.soft12019C

Views: 1987

ascl:2112.019
Add this shield to your page
Copy the above HTML to add this shield to your code's website.