kmedioids

Performs kmedioids clustering, requires only a nxn distance matrix D and number of clusters, k.

Performs k-mediods clustering; only requires a distance matrix D and number of clusters k. Finds cluster assignments "inds" to minimize the following cost function:

sum(D(inds==i,inds==i),2), summed over i=1:k

Determining cluster assignments and cluster centers are both done in an efficient, vectorized way. Cluster assignment is O(nk) and cluster centering is O(k*(max cluster size)^2)