A Keras model to classify Gravitational lensing structures. Checkout the classifier model at https://einstein-chwolson-ring-classifier.streamlit.app/
More about Dark matter substructures can be found in this paper :- https://arxiv.org/pdf/1909.07346.pdf
The model classifies the Dark matter substructure into the following three classes,
No Substructure: Gravitational Lensing Images simulated with no substructure
(no substructure)
Spherical Substructure: Gravitational Lensing Images simulated with Cold dark matter
(sphere substructure)
Vortex Substructure: Gravitational Lensing Images simulated with the superfluid dark matter
(vort substructure)
Update: The dataset was prepared from a parapet object of graviational lensing datapoints by myself using python libraries and then converted into a datset and is available on Kaggle. Kaggle notebook for training the model is "https://www.kaggle.com/code/malyala11211/gravitational-lensing-classifier"