get_cancer_data = function(){
    con = file('~/Projects/Reading/N_Z/Zelterman/Data/CANCER.DAT', open='r')
    data = c()
    while( TRUE ){
        line = readLines(con, n=1)
        line = gsub('\\?\\?', 'NA', line)
        if( length(line)==0 || nchar(line)==0 ){
            break
        }
        T = read.table(text=line, stringsAsFactors=FALSE)
        n = ncol(T)
	number_columns = 1:11
        df = as.matrix(T[, number_columns])
        if( length(data)==0 ){
            data = df
        }else{
            data = rbind(data, df)
        }
    }
    close(con)

    data = data.frame(data)
    colnames(data) = c('age', 'pop47', 'd47', ## pick names that match the names used in the book for this exercise
                              'pop51', 'd51',
        		      'pop56', 'd56',
        		      'pop61', 'd61',
        		      'pop66', 'd66'
        		      )

    data$age = as.double(data$age)
    data$pop47 = as.double(data$pop47)
    data$d47 = as.double(data$d47)
    data$pop51 = as.double(data$pop51)
    data$d51 = as.double(data$d51)
    data$pop56 = as.double(data$pop56)
    data$d56 = as.double(data$d56)
    data$pop61 = as.double(data$pop61)
    data$d61 = as.double(data$d61)
    data$pop66 = as.double(data$pop66)
    data$d66 = as.double(data$d66)

    ##DF = DF[complete.cases(DF),] ## drop an NAs
    
    return(data)
}