view release on metacpan or  search on metacpan

lib/Algorithm/MasterMind/CGA_Partitions.pm  view on Meta::CPAN

#  print "After GA combinations ", join( " ", keys %consistent ), "\n";
  $self->{'_consistent'} = \%consistent; #This mainly for outside info
  if ( $number_of_consistent > 1 ) {
#    print "Consistent ", scalar keys %consistent, "\n";
    #Use whatever we've got to compute number of partitions
    my $partitions = partitions( keys %consistent );
    
    my $max_partitions = 0;
    my %max_c;
    for my $c ( keys %$partitions ) {
      my $this_max =  keys %{$partitions->{$c}};
      $max_c{$c} = $this_max;
      if ( $this_max > $max_partitions ) {
	$max_partitions = $this_max;
      }
    }
    # Find all partitions with that max
    my @max_c = grep( $max_c{$_} == $max_partitions, keys %max_c );
    # Break ties
    my $string = $max_c[ rand( @max_c )];
    # Obtain next
    return  $self->{'_last'} = $string;
  } else {
    return $self->{'_last'} = (keys %consistent)[0];
  }
  
}

"Many blacks, 0 white"; # Magic true value required at end of module

__END__

=head1 NAME

Algorithm::MasterMind::CGA_Partitions - Solver using a Canonical GA


=head1 SYNOPSIS

    use Algorithm::MasterMind::CGA_Partitions;
    my $secret_code = 'EAFC';
    my $population_size = 200;
    my @alphabet = qw( A B C D E F );
    my $solver = new Algorithm::MasterMind::CGA_Partitions { alphabet => \@alphabet,
						length => length( $secret_code ),
						  pop_size => $population_size};
  
    #The rest, same as the other solvers

=head1 DESCRIPTION

Uses L<Algorithm::Evolutionary> instance of canonical genetic algorithm to solve MM; as there
are two different fitness functions you can use; probably
C<fitness_orig> works better. 

=head1 INTERFACE 

=head2 initialize 

Performs bookkeeping, and assigns flags depending on the
initialization values

=head2 new ( $options )

This function, and all the rest, are directly inherited from base

=head2 issue_next()

Issues the next combination

=head2 feedback()

Obtain the result to the last combination played

=head2 guesses()

Total number of guesses

=head2 evaluated()

Total number of combinations checked to issue result

=head2 number_of_rules ()

Returns the number of rules in the algorithm

=head2 rules()

Returns the rules (combinations, blacks, whites played so far) y a
reference to array

=head2 matches( $string ) 

Returns a hash with the number of matches, and whether it matches
every rule with the number of blacks and whites it obtains with each
of them

=head2 fitness( $individual )

Computes fitness summing the number of correct black and whites plus
the number of rules the combination meets times the length

=head2 fitness_orig( $individual )

Fitness proposed in the Applied and Soft Computing paper, difference
between the number of blacks/whites obtained by rules against the
secret code and by the combination against the combination in the
rule. 

=head1 SEE ALSO

Other solvers: L<Algorithm::MasterMind::Sequential> and
L<Algorithm::MasterMind::Random>. Don't work as well, really.


=head1 AUTHOR

JJ Merelo  C<< <jj@merelo.net> >>