Bernoulli Mixture Models (BMM)

This repository provides tools in a Python package bernmix for the unsupervised analysis of multivariate Bernoulli data with known number of cluster/groups using BMMs. Python 3.8.*

Maximum likelihood estimation

Shows how to fit the model using Expectation-Maximizition (EM) algorithm as outlined in Bishop (2006): Pattern Recognition and Machine Learning.

Fully Bayesian estimation

Shows how to fit the model using Gibbs sampling algorithm.

from bernmix.utils import bmm_utils as bmm

Installing

pip install -r requirements.txt

License

This project is licensed under the MIT License - see the LICENSE.md file for details