SparkR with CSV Data

gistfile1.txt

library(SparkR,lib.loc="/home/mader/tools/SparkR-pkg/lib")
sc<-sparkR.init(master="local[40]")
cfile<-textFile(sc,"/home/mader/work/simple.csv")
testinput<-take(cfile,3)
headertext<-strsplit(testinput[[1]],",")[[1]]
sample.col<-which(headertext %in% "file.name")
format.row<-function(in.row) {
  in.txt<-strsplit(in.row,",")[[1]]
  list(in.txt[sample.col],mapply(list,headertext[-sample.col],sapply(in.txt[-sample.col],as.double))[2,])
}
lab.file<-lapply(cfile,format.row)
sample.files<-groupByKey(lab.file,1000L)