Need help collapsing a list and obtaining totals in perl - arrays

Question

Hi I have a large list of data:
http://paste-it.net/public/y17027d/
It is 67859 rows by 10 columns. The 6th column contains values that represent Z-Scores from 1 to 6 in .01 increments. What I would like to do is to total all of the other column values that have the same Z-score value, but my current code is not working.
What I have now prints out values but the totals for each Z-Score are incorrect.
Here is my code:
#! /usr/bin/perl
use strict;
use warnings;
use POSIX;
use Data::Dumper;
my $input = $ARGV[0];
open (DATAFILE, $input) or die $!;
open(OUT,">>"."final.output.txt");
my($line,$fMeasure,$filename,$recall,$precision,$z_score,$computer_calls,$johns_calls,$false_negatives,$false_positives,$true_positives,$count);
$fMeasure=$filename=$recall=$precision=$z_score=$computer_calls=$johns_calls=$false_negatives=$false_positives=$true_positives=$count = 0;
my %stats=();
my %zscore=();
while($line = <DATAFILE>){
# Chop off new line character, skip the comments and empty lines.
chomp($line);
my #temp = split(/\t/, $line);
$true_positives = $temp[0];
$false_positives = $temp[1];
$false_negatives = $temp[2];
$johns_calls = $temp[3];
$computer_calls = $temp[4];
$z_score = $temp[5];
$fMeasure = $temp[6];
$precision = $temp[7];
$recall = $temp[8];
$filename = $temp[9];
$stats{$z_score}{$filename}[0] = $true_positives;
$stats{$z_score}{$filename}[1] = $false_positives;
$stats{$z_score}{$filename}[2] = $johns_calls;
$stats{$z_score}{$filename}[3] = $computer_calls;
$stats{$z_score}{$filename}[4] = $fMeasure;
$stats{$z_score}{$filename}[5] = $precision;
$stats{$z_score}{$filename}[6] = $recall;
$stats{$z_score}{$filename}[6] = $filename;
$zscore{$z_score}++;
}
my $false_positives_new = 0;
my $true_positives_new = 0;
my $johns_calls_new = 0;
my $computer_calls_new = 0;
my $file_name = 0;
foreach $z_score ( sort keys %stats ) {
foreach $filename( keys %{$stats{$z_score}} ){
my $tp = $stats{$z_score}{$filename}[0];
my $fp = $stats{$z_score}{$filename}[1];
my $jc = $stats{$z_score}{$filename}[2];
my $cc = $stats{$z_score}{$filename}[3];
my $fn = $stats{$z_score}{$filename}[6];
#print "$z_score\t$jc\n";
$false_positives_new = $false_positives + $fp;
$true_positives_new = $true_positives + $tp;
$johns_calls_new = $johns_calls + $jc;
$computer_calls_new = $computer_calls + $cc;
#print OUT "$fn\n";
}
print OUT"$true_positives_new\t$false_positives_new\t$johns_calls_new\t$computer_calls_new\t$z_score \n";
$false_positives_new = 0;
$true_positives_new = 0;
$johns_calls_new = 0;
$computer_calls_new = 0;
$file_name = 0;
}
close(OUT);
close (DATAFILE);
I know that I must be doing something wrong but I am not able to figure out what. Any help would be greatly appreciated. Thank you

Answer 1

OK. I was able to get the data from pastebin and I think the following code does what you want.
#! /usr/bin/perl
use strict; use warnings;
use Data::Dumper;
my ($input) = #ARGV;
open my $DATAFILE, '<', $input
or die "Cannot open '$input': $!";
my #field_names = qw(
fMeasure
recall
precision
z_score
computer_calls
johns_calls
false_negatives
false_positives
true_positives
count
);
my #track_fields = qw(
false_positives
false_negatives
johns_calls
computer_calls
);
my (%stats, %by_zscore);
while ( my $line = <$DATAFILE> ) {
last unless $line =~ /\S/;
chomp $line;
my #temp = split /\t/, $line;
my $filename = pop #temp;
my %fields;
#fields{ #field_names } = #temp;
my $z_score = $fields{z_score};
$stats{ $z_score }{$filename} = \#temp;
for my $f ( #track_fields ) {
$by_zscore{$z_score}{ $f } += $fields{ $f };
}
}
print Dumper \%by_zscore;

Answer 2

I think you want to say
$false_positives_new = $false_positives_new + $fp;
# etc.
instead of
$false_positives_new = $false_positives + $fp;