Clustering

This crate provides an easy and efficient way to perform kmeans clustering on arbitrary data. The algo is initialized with kmeans++ for best performance of the clustering.

There are three goals to this implementation of the kmeans algorithm:

1. it must be generic
2. it must be easy to use
3. it must be reasonably fast

Example

```rust use clustering::*;

let nsamples = 20000; // # of samples in the example let ndimensions = 200; // # of dimensions in each sample let k = 4; // # of clusters in the result let maxiter = 100; // max number of iterations before the clustering forcefully stops

// Generate some random data let mut samples: Vec> = vec![]; for _ in 0..nsamples { samples.push((0..ndimensions).map(|_| rand::random()).collect::>()); }

// actually perform the clustering let clustering = kmeans(k, &samples, max_iter);

println!("membership: {:?}", clustering.membership); println!("centroids : {:?}", clustering.centroids); ```