Package provides an easy, fast and scalable way to perform inference in Dirichlet Process Mixture Models.


Developed from the code of:

Distributed MCMC Inference in Dirichlet Process Mixture Models Using Julia by Dinari et al.

Which is based on the algorithm from:

Parallel Sampling of DP Mixture Models using Sub-Clusters Splits by Chang and Fisher.

The package currently supports Gaussian and Multinomial priors, however adding your own is very easy, and more will come in future releases.


2d Gaussian with plotting

Image Segmentation

Example of running from a params file, including saving and loading

If you use this package in your research, please cite the following:

  title={Distributed MCMC Inference in Dirichlet Process Mixture Models Using Julia},
  author={Dinari, Or and Yu, Angel and Freifeld, Oren and Fisher III, John W},
  booktitle={2019 19th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)},

For any questions: dinari@post.bgu.ac.il Also available on Julia's Slack.

Contributions, feature requests, suggestion etc.. are welcomed.