Help with Histograms

[cmr72]

Hello All, I am a newbie. I've been a round PERL for a while but never had to really get down to the nitty gritty of it. I am not a programmer and usually troll around looking for answers. Today I created a random number generator in hopes to pipe the output into a histogram set for plotting in GNUPLOT. While I understand how to generate a hash to tally the points, I am totally lost on creating bins, whether automatically spaced or with an argument from the user. Here is the code that I have so far:

Random number generator:

#!/usr/bin/perl -w

use strict ;

my $samplesize = $ARGV[0] ;
my $max        = $ARGV[1] ;

chomp(my $prog = `basename $0`) ;

if ($#ARGV < 1)
{
  print "\nUsage: ./randomnumber.pl <samplesize> <max num>\n" ;
  exit ;
}


for (my $i = 0; $i<$samplesize; $i++)
{
  my $range = $max ;
  my $random_number = int(rand($range)) - $max/2;
  print "$random_number \n";
}

Histogram creator:

#!/usr/bin/perl -w

use strict ;


my %bin ;
my %size;
my %value ;
my $max = 0 ;
my $min = 0 ;
my $count ;


while ( <STDIN> ) 
{
   my @line= split  ;
   my $element = $line[0] ;
   $value{$element}++ ;
}



foreach my $i (sort {$a<=>$b} keys %value)
{
  if ($i > $max)
  {
    $max = $i ;
  }
  if ($i < $min)
  {
    $min = $i ;
  }
}

So if there are values -1, -1.5, -2, 2, 1, 0, 1, I would hope to see the following (if the bin range was between -2:2)

(-2) 3

(0) 1

(2) 3

Or something like that... Thanks in advance.

[toolic]

Look at how others have created histograms (or just use their code):
• Super Search for histogram
• CPAN search...
• Text::Histogram
• Text::HistogramChart

[philiprbrenan]

Given that the input is a sequence of random numbers, you could synthesize your desired arrangement of bins directly with no loss of information:

use feature ":5.14";
use warnings FATAL => qw(all);
use strict;
use Data::Dump qw(dump pp);

sub bins($$$$)# Fill some bins
 {my ($start, $count, $width, $samples) = @_; # Start bin, number of b
+ins, bin width, number of samples
  my $bins;
  $bins->{$start+$width * int(rand $count)}++ for 1..$samples;
  $bins
 }

pp(bins(-2, 3, 2, 8));

Produces:

{ "-2" => 4, "0" => 1, "2" => 3 }

[pvaldes]

#---------------------------
## hi, I am a gnuplot script and you are probably trying to do in perl
+ my own job

set terminal png large size 1920,1200
set output 'output.png'
set style histogram
set title "my title"
set xlabel "my x varname"
set ylabel "my y varname"
set yrange [0:10]
set style fill solid border -1

binwidth=5
set boxwidth binwidth

bin(x,width)=width*floor(x/width) + binwidth/2.0

plot 'mydata.dat' using (bin($1,binwidth)):(1.0) smooth freq with boxe
+s 
#---------------------------

[cmr72]

Hey all, thanks for your help. I managed to do the following:

#!/usr/bin/perl -w
use strict ;
use POSIX ;

my %bin ;
my %size;
my %value ;
my $binwidth = $ARGV[0] ;

chomp(my $prog = `basename $0`) ;

if ($#ARGV < 0)
{
  print "\nUsage: ./histo.pl.pl <bin width>\n" ;
  exit ;
}



while ( <STDIN> ) {
   my @line= split  ;
   my $element = $line[0] ;
   $value{ceil(($element+1)/$binwidth)-1}++
}



foreach my $i (sort {$a<=>$b} keys %value)
{
   my $bin = $i*$binwidth ;
   print "$bin $value{$i}\n" ;
}