R/ClassifyByDecisionBoundaries.R

In Mthrun/AdaptGauss: Gaussian Mixture Models (GMM)

ClassifyByDecisionBoundaries=function(Data,DecisionBoundaries,ClassLabels){
# Cls = ClassifyByDecisionBoundaries(Data,DecisionBoundaries)
# Classify Data according to decision Boundaries
#
# INPUT
# Data(1:n,1)                 vector of Data
# DecisionBoundaries(1:L)    decision boundaries 
# OPTIONAL
# ClassLabels(1:L+1)        numbered class labels that are assigned to the classes. default (1:L) 
# OUTPUT
# Cls(1:n,1:d)               classiffication of Data
# Author MT 04/2015


if(!is.vector(Data)){
  warning('Data converted to vector')
 Data=as.vector(Data) 
}
if(is.list(DecisionBoundaries)){
  DecisionBoundaries=as.vector(DecisionBoundaries$DecisionBoundaries)
  print('DecisionBoundaries was a list, assuming usage of BayesDecisionBoundaries()')
}
AnzBounds= length(DecisionBoundaries)
if(missing(ClassLabels)){
ClassLabels=seq(from=1,by=1,to=(AnzBounds+1))
}

Cls=rep(1,length(Data)) # default alles in Klasse 1

nonan=which(is.finite(Data))

if(length(nonan)!=length(Data)){
  warning('Datavector contains NaN. These values cannot be classified.')
  names(Cls)=1:length(Data)
  ClsTmp=Cls[nonan]
  DataTmp=Data[nonan]
  for(b in 1:AnzBounds){
    ind=DataTmp>DecisionBoundaries[b]
    ClsTmp[ind] = rep(ClassLabels[b+1],sum(ind))
  } # for c
  Cls[as.numeric(names(ClsTmp))]=ClsTmp
  Cls[setdiff(1:length(Cls),as.numeric(names(ClsTmp)))]=NaN
}else{
  for(b in 1:AnzBounds){
    ind=Data>DecisionBoundaries[b]
    Cls[ind] = rep(ClassLabels[b+1],sum(ind))
  } # for c
}
return(Cls)
}