Classification of pulsars using Extreme Deconvolution

We carry out a classification of the observed pulsar dataset into distinct clusters, based on $P-\dot{P}$ diagram, using Extreme Deconvolution Gaussian Mixture Model. then use Bayesian Information Criterion to select optimum number clusters. find in accord with previous works, that can be optimally classified six two for millisecond population, and four ordinary population. Beyond that, however we do not glean any additional insight population this classification. Using numerical experiments, confirm Deconvolution-based is less sensitive variations compared Models. All our analysis codes used work have been made publicly available.