R/predict.r

In biclust: BiCluster Algorithms

predictBimax<-function(BCrepBimax,x)
{
le<-dim(x)[1]
count<-BCrepBimax@Number
coln<-rowSums(BCrepBimax@NumberxCol)
res<-vector("integer",length=le)
count
for(i in 1:le)
  {
  xtest<-x[i,]
  for(j in 1:count)
    {
    if(sum(xtest[BCrepBimax@NumberxCol[j,]])==coln[j])
      {
      res[i]<-j
      break
      }      
    }
  }

#res[is.na(res)]<-0
res
}