In this example the k-means clustering method is used to cluster a given toy
data set. In k-means clustering one tries to partition n observations into k
clusters in which each observation belongs to the cluster with the nearest mean.
The algorithm class constructor takes the number of clusters and a distance to
be used as input. The distance used in this example is Euclidean distance.
After training one can fetch the result of clustering by obtaining the cluster
centers and their radiuses.
