Selection of k in k-means clustering based on Pham et al. paper ``Selection of k in k-means clustering''.
kselectionallows to inject a function that provides a different implementation of kmeans algorithm. This feature has developed thanks Harold Pimentel.
nstartis not a parameter of
kselectionto facilitate compatibility with different implementations of kmeans algorithm.
kselectionallows the parallel execution of the algorithm using
traceoption in the
kselectionfunction shows a trace, the progres must be selected using the
kselectionfunction display a trace of the progress.
kselectionwhen the number of rows in the input data is lower than 15.