Skip to content

dturpin-svom/otrain

Repository files navigation

O'TRAIN: Optical TRAnsient Identification NEtwork: Easy!

Development status

Build Status codecov Documentation Status

Generic Tools to help identification of transients in astronomical images, based on Convolutional Neural Network.

Aim

  • Implement a general tool based on machine learning to help identifying transient in astonmical images.
  • Effortless execution through the command line.
  • Usable on different kind of images from different telescopes.
  • OS-independent, so that it can work on Windows, Mac OS and Linux machines.

Features

  • Usage of Docker to allow deployment on different OS, container based on Ubuntu 20.04.
  • User provide image cutouts containing real and false transients.
  • These cutouts are used to train a CNN algorithm implemented with Keras.
  • Built-in diagnostics help to characterise accuracy of training.
  • Usage of trained model to classify any new cutouts

Installation

See documentation: https://otrain.readthedocs.io.

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published