pyrichlet: A Python Package for Density Estimation and Clustering Using Gaussian Mixture Models

pyrichlet: A Python Package for Density Estimation and Clustering Using Gaussian Mixture Models

Bayesian nonparametric models have proven to be successful tools for clustering and density estimation. While there exists a nourished ecosystem of implementations in R, for Python there are only a few. Here we develop a Python package called pyrichlet, for Bayesian nonparametric density estimation and clustering using various state-of-the-art Gaussian mixture models that generalize the well established Dirichlet process mixture, many of which are fairly new. Implementation is performed using Markov chain Monte Carlo techniques as well as variational Bayes methods. This article contains a detailed description of pyrichlet and examples for its usage with a real dataset.

Bayesian nonparametric models have proven to be successful tools for clustering and density estimation. While there exists a nourished ecosystem of implementations in R, for Python there are only a few. Here we develop a Python package called pyrichlet, for Bayesian nonparametric density estimation and clustering using various state-of-the-art Gaussian mixture models that generalize the well established Dirichlet process mixture, many of which are fairly new. Implementation is performed using Markov chain Monte Carlo techniques as well as variational Bayes methods. This article contains a detailed description of pyrichlet and examples for its usage with a real dataset.

Fidel Selva, Ruth Fuentes-García, María Fernanda Gil-Leyva

https://www.jstatsoft.org/article/view/v112i08

OJS/PKP

29 MARET 2025

FAJAR BAGUS W

PDF

ENGLISH

TEXT