Skip to main content

Unsupervised and Semi-supervised Dirichlet Process Heterogeneous Mixtures

Project description

Implements Dirichlet Process Heterogeneous Mixtures of exponential family distributions for clustering heterogeneous data without choosing the number of clusters. Inference can be performed with Gibbs sampling or coordinate ascent mean-field variational inference. For semi-supervised learning, Gibbs sampling supports must-link and cannot-link constraints. A novel variational inference algorithm was derived to handle must-link constraints.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dphmix-0.2.0.tar.gz (15.5 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page