% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/HelperFuncs.R
\name{estimClustNum}
\alias{estimClustNum}
\title{Wrapper for estimation of cluster number}
\usage{
estimClustNum(dat, maxClust = 25, cores = 1)
}
\arguments{
\item{dat}{matrix of features averaged over replicates. The last column contains their standard deviation}

\item{maxClust}{Maximal number of cluster. The minimum is 3}

\item{cores}{The number of threads to be used for parallelisation}
}
\value{
list with the items `ClustInd`: list of clustering objects for each number of clusters, `p` plot object with plots for validity indices, `numclust` optimal cluster number according to "minimum centroid distance"
}
\description{
This runs the clustering for different numbers of clusters, and estimates the most suitable numbers from applying
the minimum centroid distance and the Xie Beni index. Multi-threading is used to shorten the computation times. 
Given the hierarchical structure of many data sets, the resulting 
numbers are suggestions. Inspection of the here plotted indices help to determine alternative cluster numbers, 
given by a strong decay of the minimum centroid distance and/or a low value of the Xie Beni index.
}
\examples{
data <- matrix(rnorm(1000), nrow=100)
estim_out <- estimClustNum(data, maxClust=10)
best_number <- max(estim_out[1])
}