kmedoids

Source: src/kmedoids.rs

K-Medoids (PAM) clustering with precomputed distance matrix.

Accepts a flat distance matrix and runs the Partitioning Around Medoids swap algorithm in Rust, returning cluster assignments.

Python API

kmedoids_pam

kmedoids_pam(dist_flat, n, k, max_iter=100, seed=42)

Rust-accelerated k-medoids PAM algorithm.

Accepts a flat distance matrix (list of n*n floats, row-major), the number of series n, clusters k, max iterations, and seed. Returns a list of cluster assignments (0-indexed) for each series.

Internal Functions

These are internal Rust functions not directly callable from Python.

pam_swap

Run the PAM swap algorithm on a precomputed distance matrix.

Arguments

Parameter	Description
dist_flat	Flattened n×n distance matrix (row-major)
n	Number of series
k	Number of clusters
max_iter	Maximum swap iterations
seed	Random seed for initial medoid selection

Returns (medoid_indices, cluster_assignments) as Vec.