Re: largest number inside array/hash

You could always just sort it and then take the highest one. I'm not sure how that compares speed wise with just scanning threw 1 by 1. Of course you could just keep a $counter and increment it when adding and not decrement it when removing. Then there is no need to find the highest because you remember it.

my $highest;
for (@array) { # or keys %hash
    $highest = $_ if $_ > $highest;
}
[download]

___________
Eric Hodges

Comment on Re: largest number inside array/hash Download Code

Replies are listed 'Best First'.

Re: Re: largest number inside array/hash
by eric256 (Parson) on Apr 06, 2004 at 20:46 UTC

I did some quick playing with the methods shown here just to get an idea of speed. Before now i've never looked at List::Util because I could do it all myself but now I think I see the light. It provides an astounding speed bonus.

use strict;
use warnings;
use Benchmark ;
use List::Util qw(max shuffle);

   
my @a = (1..500_000);
my @b = shuffle (1..500_000);
my $greatest;

timethese(25, {
     'sort_shuffled' =>
          sub {($greatest)=sort{$b<=>$a}@b;},
     'grep_shuffled' =>   
          sub { 
              grep($greatest=($_>$greatest)?$_:$greatest,@b); 
          },
     'for_shuffled' => 
          sub {
              $greatest = 0;
              for (@b) { $greatest = $_ if $_ > $greatest; }
          },
     'max_shuffled' => 
           sub { max(@b) },
     'sort_inorder' =>
          sub {($greatest)=sort{$b<=>$a}@a;},
     'grep_inorder' =>   
          sub { 
              grep($greatest=($_>$greatest)?$_:$greatest,@a); 
          },
     'for_inorder' => 
          sub {
              $greatest = 0;
              for (@a) { $greatest = $_ if $_ > $greatest; }
          },
     'max_inorder' => 
           sub { max(@a) }
});

__DATA__
Benchmark: timing 25 iterations of for_inorder, for_shuffled, grep_ino
+rder, grep_shuffled, max_inord
er, max_shuffled, sort_inorder, sort_shuffled...
for_inorder:  5 wallclock secs ( 5.17 usr +  0.00 sys =  5.17 CPU) @  
+4.83/s (n=25)
for_shuffled:  4 wallclock secs ( 3.55 usr +  0.00 sys =  3.55 CPU) @ 
+ 7.05/s (n=25)
grep_inorder:  4 wallclock secs ( 3.75 usr +  0.00 sys =  3.75 CPU) @ 
+ 6.67/s (n=25)
grep_shuffled:  4 wallclock secs ( 3.70 usr +  0.00 sys =  3.70 CPU) @
+  6.75/s (n=25)
max_inorder:  1 wallclock secs ( 0.89 usr +  0.00 sys =  0.89 CPU) @ 2
+8.09/s (n=25)
max_shuffled:  1 wallclock secs ( 0.91 usr +  0.00 sys =  0.91 CPU) @ 
+27.59/s (n=25)
sort_inorder:  1 wallclock secs ( 1.73 usr +  0.02 sys =  1.75 CPU) @ 
+14.28/s (n=25)
sort_shuffled: 36 wallclock secs (34.34 usr +  0.05 sys = 34.39 CPU) @
+  0.73/s (n=25)
[download]

I was curious if anyone could explain why grep is so much faster thant the for? Ohh and how do people do those cool comparison benchmarks with the chart?

___________
Eric Hodges

[reply]
[d/l]

Re: Re: Re: largest number inside array/hash

by PodMaster (Abbot) on Apr 06, 2004 at 21:47 UTC

Ohh and how do people do those cool comparison benchmarks with the chart?

MJD says "you can't just make shit up and expect the computer to know what you mean, retardo!"
I run a Win32 PPM repository for perl 5.6.x and 5.8.x -- I take requests (README).
** The third rule of perl club is a statement of fact: pod is sexy.

[reply]


Problems? Is your data what you think it is?
	PerlMonks