KCenters.jl

The KCenters package implements algorithms that approximate the kcenters problem, in particular, we use farthest first traversal and density nets. It adds several variants of center selection and stopping criterions, along with a number of heuristics and utilities for taking advantage of the resulting groups.

• The majority of tasks will use kcenters that outputs a ClusteringData.
• Low level tasks may require to use enet or dnet.
• Partitioning data differently to ClusteringData.
• For more detailed usage, please visit each one of the pages on the left panel.