Crowdastro was my honours project at the Australian National University, supervised by Dr Cheng Soon Ong and Dr Julie Banfield. We used crowdsourced astronomy data from Radio Galaxy Zoo to train computers to cross-identify active galactic nuclei with their host galaxies.
A major outcome of this project was the development of an open source Python package for machine learning on radio survey images, called Crowdastro. Crowdastro is MIT-licensed and is available on PyPI or GitHub. Documentation is available on Read the Docs.
You can read my final presentation slides here. My thesis is available here.