dbscan

Dependency free implementation of DBSCAN clustering in Rust, generic over the numeric types from Rust's stdlib.

lib.rs:

A Density-Based Algorithm for Discovering Clusters

This algorithm finds all points within eps distance of each other and attempts to cluster them. If there are at least mpt points reachable (within distance eps) from a given point P, then all reachable points are clustered together. The algorithm then attempts to expand the cluster, finding all border points reachable from each point in the cluster

See Ester, Martin, et al. "A density-based algorithm for discovering clusters in large spatial databases with noise." Kdd. Vol. 96. No. 34. 1996. for the original paper

Thanks to the rusty_machine implementation for inspiration