AI-Genetic-Pro
view release on metacpan or search on metacpan
0.34 Tue, 17 Mar 2009 20:39:16 +0100
- Fixed bug in PMX strategy. Thanks to Maciej Misiak :-)
0.335 Sat, 07 Feb 2009 20:04:52 +0100
- Little changes in a Makefile.PL (especially for Sun Solaris)
0.334 Fri, 23 Jan 2009 00:03:26 +0100
- Module 'Digest::MD5' is loaded by default,
0.333 Fri, 22 Jan 2009 15:30:06 +0100
- Some improvments in 'getFittest' function,
- Added 'getFittest_as_arrayref' function,
0.332 Wed, 21 Jan 2009 00:31:01 +0100
- Some changes in tests.
0.331 Tue, 20 Jan 2009 23:55:20 +0100
- Added tests.
- Some improvments in the 'inject' function.
0.33 Mon, 19 Jan 2009 00:38:06 +0100
- Added 'strict' mode.
- Added 'inject' function.
0.32 Sun, 18 Jan 2009 01:16:37 +0100
- Fixes in rangevectors.
0.31 Sat, 17 Jan 2009 20:32:27 +0100
sections when you distribute them as separate
works. But when you distribute the same
sections as part of a whole which is a work
based on the Library, the distribution of the
whole must be on the terms of this License,
whose permissions for other licensees extend
to the entire whole, and thus to each and every
part regardless of who wrote it.
Thus, it is not the intent of this section to claim
rights or contest your rights to work written
entirely by you; rather, the intent is to exercise
the right to control the distribution of derivative
or collective works based on the Library.
In addition, mere aggregation of another work
not based on the Library with the Library (or
with a work based on the Library) on a volume
of a storage or distribution medium does not
bring the other work under the scope of this
License.
indirectly through you, then the only way you could satisfy
both it and this License would be to refrain entirely from
distribution of the Library.
If any portion of this section is held invalid or unenforceable
under any particular circumstance, the balance of the
section is intended to apply, and the section as a whole is
intended to apply in other circumstances.
It is not the purpose of this section to induce you to infringe
any patents or other property right claims or to contest
validity of any such claims; this section has the sole purpose
of protecting the integrity of the free software distribution
system which is implemented by public license practices.
Many people have made generous contributions to the wide
range of software distributed through that system in reliance
on consistent application of that system; it is up to the
author/donor to decide if he or she is willing to distribute
software through any other system and a licensee cannot
impose that choice.
t/04_bitvectors_variable_length_I.t
t/05_bitvectors_variable_length_II.t
t/06_listvectors_constant_length.t
t/07_listvectors_variable_length_I.t
t/08_listvectors_variable_length_II.t
t/09_rangevectors_constant_length.t
t/10_rangevectors_variable_length_I.t
t/11_rangevectors_variable_length_II.t
t/12_combinations_constant_length.t
t/13_preserve.t
t/14_getFittest.t
t/15_bitvectors_constant_length_MCE.t
t/16_bitvectors_constant_length_-_native_arrays.t
t/17_bitvectors_constant_length_MCE_-_native_arrays.t
Makefile.PL view on Meta::CPAN
"List::Util" => 0,
"MCE" => "1.874",
"MCE::Map" => "1.874",
"Math::Random" => "0.72",
"Storable" => "2.05",
"Struct::Compare" => 0,
"Tie::Array::Packed" => "0.13",
"UNIVERSAL::require" => 0
},
"VERSION" => "1.009",
"test" => {
"TESTS" => "t/*.t"
}
);
my %FallbackPrereqs = (
"Carp" => 0,
"Class::Accessor::Fast::XS" => 0,
"Clone" => 0,
"Digest::MD5" => 0,
use AI::Genetic::Pro;
sub fitness {
my ($ga, $chromosome) = @_;
return oct('0b' . $ga->as_string($chromosome));
}
sub terminate {
my ($ga) = @_;
my $result = oct('0b' . $ga->as_string($ga->getFittest));
return $result == 4294967295 ? 1 : 0;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 1000, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.01, # probab. of mutation
-workers => 3, # number of workers (MCE)
);
# init population of 32-bit vectors
$ga->init(32);
# evolve 10 generations
$ga->evolve(10);
# best score
print "SCORE: ", $ga->as_value($ga->getFittest), ".\n";
# save evolution path as a chart
$ga->chart(-filename => 'evolution.png');
# save state of GA
$ga->save('genetic.sga');
# load state of GA
$ga->load('genetic.sga');
changes). Additionally AI::Genetic::Pro isn't a pure Perl solution, so
it doesn't have limitations of its ancestor (such as slow-down in the
case of big populations ( >10000 ) or vectors with more than 33
fields).
If You are looking for a pure Perl solution, consider AI::Genetic.
Speed
To increase speed XS code is used, however with portability in mind.
This distribution was tested on Windows and Linux platforms (and
should work on any other).
Multicore support is available through Many-Core Engine (MCE). You
can gain the most speed up for big populations or time/CPU consuming
fitness functions, however for small populations and/or simple
fitness function better choice will be single-process version.
You can get even more speed up if you turn on use of native arrays
(parameter: native) instead of packing chromosomes into single
scalar. However you have to remember about expensive memory use in
This option has any meaning only if MCE is turned on. This defines
how many process will be used during processing. Default will be
used one proces per core (most efficient).
-strict
This defines if the check for modifying chromosomes in a
user-defined fitness function is active. Directly modifying
chromosomes is not allowed and it is a highway to big trouble. This
mode should be used only for testing, because it is slow.
$ga->inject($chromosomes)
Inject new, user defined, chromosomes into the current population.
See example below:
# example for bitvector
my $chromosomes = [
[ 1, 1, 0, 1, 0, 1 ],
[ 0, 0, 0, 1, 0, 1 ],
[ mean, mean1, mean2, ... ], # mean values
[ min0, min1, min2, ... ], # min values
]
$ga->getAvgFitness()
Get max, mean and min score of the current generation. In example:
my ($max, $mean, $min) = $ga->getAvgFitness();
$ga->getFittest($n, $unique)
This function returns a list of the fittest chromosomes from the
current population. You can specify how many chromosomes should be
returned and if the returned chromosomes should be unique. See
example below.
# only one - the best
my ($best) = $ga->getFittest;
# or 5 bests chromosomes, NOT unique
my @bests = $ga->getFittest(5);
# or 7 bests and UNIQUE chromosomes
my @bests = $ga->getFittest(7, 1);
If you want to get a large number of chromosomes, try to use the
getFittest_as_arrayref function instead (for efficiency).
$ga->getFittest_as_arrayref($n, $unique)
This function is very similar to getFittest, but it returns a
reference to an array instead of a list.
$ga->generation()
Get the number of the current generation.
$ga->people()
Returns an anonymous list of individuals/chromosomes of the current
population.
SUPPORT
AI::Genetic::Pro is still under development; however, it is used in
many production environments.
TODO
Examples.
More tests.
More warnings about incorrect parameters.
REPORTING BUGS
When reporting bugs/problems please include as much information as
possible. It may be difficult for me to reproduce the problem as almost
every setup is different.
A small script which yields the problem will probably be of help.
Tod Hagan for reporting a bug (rangevector values truncated to signed
8-bit quantities) and supplying a patch.
Randal L. Schwartz for reporting a bug in this documentation.
Maciej Misiak for reporting problems with combination (and a bug in a
PMX strategy).
LEONID ZAMDBORG for recommending the addition of variable-length
chromosomes as well as supplying relevant code samples, for testing and
at the end reporting some bugs.
Christoph Meissner for reporting a bug.
Alec Chen for reporting some bugs.
AUTHOR
Strzelecki Lukasz <lukasz@strzeleccy.eu>
lib/AI/Genetic/Pro.pm view on Meta::CPAN
my @preserved;
for(my $i = 0; $i != $generations; $i++){
# terminate ----------------------------------------------------
last if $self->terminate and $self->terminate->($self);
# update generation --------------------------------------------
$self->generation($self->generation + 1);
# update history -----------------------------------------------
$self->_save_history;
#---------------------------------------------------------------
# preservation of N unique chromosomes
@preserved = map { clone($_) } @{ $self->getFittest_as_arrayref($self->preserve - 1, 1) };
# selection ----------------------------------------------------
$self->_select_parents();
# crossover ----------------------------------------------------
$self->_crossover();
# mutation -----------------------------------------------------
$self->_mutation();
#---------------------------------------------------------------
for(@preserved){
my $idx = int rand @{$self->chromosomes};
$self->chromosomes->[$idx] = $_;
lib/AI/Genetic/Pro.pm view on Meta::CPAN
sub getHistory { $_[0]->_history() }
#=======================================================================
sub mutProb { shift->mutation(@_) }
#=======================================================================
sub crossProb { shift->crossover(@_) }
#=======================================================================
sub intType { shift->type() }
#=======================================================================
# STATS ################################################################
#=======================================================================
sub getFittest_as_arrayref {
my ($self, $n, $uniq) = @_;
$n ||= 1;
$self->_calculate_fitness_all() unless scalar %{ $self->_fitness };
my @keys = sort { $self->_fitness->{$a} <=> $self->_fitness->{$b} } 0..$#{$self->chromosomes};
if($uniq){
my %grep;
my $chromosomes = $self->chromosomes;
if( my $pkg = $self->_package ){
lib/AI/Genetic/Pro.pm view on Meta::CPAN
my $key = md5_hex( join( q[:], @{ $chromosomes->[ $_ ] } ) );
$tmp{ $key } && 0 or $tmp{ $key } = 1;
} @keys;
}
}
$n = scalar @keys if $n > scalar @keys;
return [ reverse @{$self->chromosomes}[ splice @keys, $#keys - $n + 1, $n ] ];
}
#=======================================================================
sub getFittest { return wantarray ? @{ shift->getFittest_as_arrayref(@_) } : shift @{ shift->getFittest_as_arrayref(@_) }; }
#=======================================================================
sub getAvgFitness {
my ($self) = @_;
my @minmax = minmax values %{$self->_fitness};
my $mean = sum(values %{$self->_fitness}) / scalar values %{$self->_fitness};
return $minmax[1], int($mean), $minmax[0];
}
#=======================================================================
1;
lib/AI/Genetic/Pro.pm view on Meta::CPAN
use AI::Genetic::Pro;
sub fitness {
my ($ga, $chromosome) = @_;
return oct('0b' . $ga->as_string($chromosome));
}
sub terminate {
my ($ga) = @_;
my $result = oct('0b' . $ga->as_string($ga->getFittest));
return $result == 4294967295 ? 1 : 0;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 1000, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.01, # probab. of mutation
lib/AI/Genetic/Pro.pm view on Meta::CPAN
-workers => 3, # number of workers (MCE)
);
# init population of 32-bit vectors
$ga->init(32);
# evolve 10 generations
$ga->evolve(10);
# best score
print "SCORE: ", $ga->as_value($ga->getFittest), ".\n";
# save evolution path as a chart
$ga->chart(-filename => 'evolution.png');
# save state of GA
$ga->save('genetic.sga');
# load state of GA
$ga->load('genetic.sga');
lib/AI/Genetic/Pro.pm view on Meta::CPAN
doesn't have limitations of its ancestor (such as slow-down in the
case of big populations ( >10000 ) or vectors with more than 33 fields).
If You are looking for a pure Perl solution, consider L<AI::Genetic>.
=over 4
=item Speed
To increase speed XS code is used, however with portability in
mind. This distribution was tested on Windows and Linux platforms
(and should work on any other).
Multicore support is available through Many-Core Engine (C<MCE>).
You can gain the most speed up for big populations or time/CPU consuming
fitness functions, however for small populations and/or simple fitness
function better choice will be single-process version.
You can get even more speed up if you turn on use of native arrays
(parameter: C<native>) instead of packing chromosomes into single scalar.
However you have to remember about expensive memory use in that case.
lib/AI/Genetic/Pro.pm view on Meta::CPAN
=item -workers
This option has any meaning only if MCE is turned on. This defines how
many process will be used during processing. Default will be used one proces per core (most efficient).
=item -strict
This defines if the check for modifying chromosomes in a user-defined fitness
function is active. Directly modifying chromosomes is not allowed and it is
a highway to big trouble. This mode should be used only for testing, because it is B<slow>.
=back
=item I<$ga>-E<gt>B<inject>($chromosomes)
Inject new, user defined, chromosomes into the current population. See example below:
# example for bitvector
my $chromosomes = [
[ 1, 1, 0, 1, 0, 1 ],
lib/AI/Genetic/Pro.pm view on Meta::CPAN
[ mean, mean1, mean2, ... ], # mean values
[ min0, min1, min2, ... ], # min values
]
=item I<$ga>-E<gt>B<getAvgFitness>()
Get I<max>, I<mean> and I<min> score of the current generation. In example:
my ($max, $mean, $min) = $ga->getAvgFitness();
=item I<$ga>-E<gt>B<getFittest>($n, $unique)
This function returns a list of the fittest chromosomes from the current
population. You can specify how many chromosomes should be returned and if
the returned chromosomes should be unique. See example below.
# only one - the best
my ($best) = $ga->getFittest;
# or 5 bests chromosomes, NOT unique
my @bests = $ga->getFittest(5);
# or 7 bests and UNIQUE chromosomes
my @bests = $ga->getFittest(7, 1);
If you want to get a large number of chromosomes, try to use the
C<getFittest_as_arrayref> function instead (for efficiency).
=item I<$ga>-E<gt>B<getFittest_as_arrayref>($n, $unique)
This function is very similar to C<getFittest>, but it returns a reference
to an array instead of a list.
=item I<$ga>-E<gt>B<generation>()
Get the number of the current generation.
=item I<$ga>-E<gt>B<people>()
Returns an anonymous list of individuals/chromosomes of the current population.
lib/AI/Genetic/Pro.pm view on Meta::CPAN
C<AI::Genetic::Pro> is still under development; however, it is used in many
production environments.
=head1 TODO
=over 4
=item Examples.
=item More tests.
=item More warnings about incorrect parameters.
=back
=head1 REPORTING BUGS
When reporting bugs/problems please include as much information as possible.
It may be difficult for me to reproduce the problem as almost every setup
is different.
lib/AI/Genetic/Pro.pm view on Meta::CPAN
Miles Gould for suggestions and some fixes (even in this documentation! :-).
Alun Jones for fixing memory leaks.
Tod Hagan for reporting a bug (rangevector values truncated to signed 8-bit quantities) and supplying a patch.
Randal L. Schwartz for reporting a bug in this documentation.
Maciej Misiak for reporting problems with C<combination> (and a bug in a PMX strategy).
LEONID ZAMDBORG for recommending the addition of variable-length chromosomes as well as supplying relevant code samples, for testing and at the end reporting some bugs.
Christoph Meissner for reporting a bug.
Alec Chen for reporting some bugs.
=head1 AUTHOR
Strzelecki Lukasz <lukasz@strzeleccy.eu>
=head1 SEE ALSO
lib/AI/Genetic/Pro/Chromosome.pm view on Meta::CPAN
use Tie::Array::Packed;
#use Math::Random qw(random_uniform_integer);
#=======================================================================
sub new {
my ($class, $data, $type, $package, $length) = @_;
my @genes;
tie @genes, $package if $package;
if($type eq q/bitvector/){
#@genes = random_uniform_integer(scalar @$data, 0, 1); # this is fastest, but uses more memory
@genes = map { rand > 0.5 ? 1 : 0 } 0..$length; # this is faster
#@genes = split(q//, unpack("b*", rand 99999), $#$data + 1); # slow
}elsif($type eq q/combination/){
#@genes = shuffle 0..$#{$data->[0]};
@genes = shuffle 0..$length;
}elsif($type eq q/rangevector/){
@genes = map { $_->[1] + int rand($_->[2] - $_->[1] + 1) } @$data[0..$length];
}else{
@genes = map { 1 + int(rand( $#{ $data->[$_] })) } 0..$length;
}
t/01_inject.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/03_bitvectors_constant_length.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/03_bitvectors_constant_length.t view on Meta::CPAN
[ qw( 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/04_bitvectors_variable_length_I.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/04_bitvectors_variable_length_I.t view on Meta::CPAN
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/05_bitvectors_variable_length_II.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/05_bitvectors_variable_length_II.t view on Meta::CPAN
[ qw( 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/06_listvectors_constant_length.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'listvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/06_listvectors_constant_length.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/07_listvectors_variable_length_I.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'listvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/07_listvectors_variable_length_I.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/08_listvectors_variable_length_II.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'listvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.01, # probab. of mutation
t/08_listvectors_variable_length_II.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/09_rangevectors_constant_length.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'rangevector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/09_rangevectors_constant_length.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/10_rangevectors_variable_length_I.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'rangevector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/10_rangevectors_variable_length_I.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/11_rangevectors_variable_length_II.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'rangevector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/11_rangevectors_variable_length_II.t view on Meta::CPAN
[qw( 0 4 0 4 0 4 0 4 )],
[qw( 4 4 0 0 4 4 0 0 )],
[qw( 4 4 4 4 0 0 0 0 )],
[qw( 0 0 0 0 4 4 4 4 )],
);
push @data, @data for 1..SIZE;
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/12_combinations_constant_length.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return calc(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'combination', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/12_combinations_constant_length.t view on Meta::CPAN
[qw( a b d c e f h g )],
[qw( a c b d f e g h )],
[qw( h b c d e f g a )],
);
push @data, @data for 1..scalar(@Win);
$ga->inject(\@data);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/13_preserve.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/14_getFittest.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/14_getFittest.t view on Meta::CPAN
# init population of 32-bit vectors
$ga->init(BITS);
$ga->inject( [ \@Win, \@Win, \@Win, \@Win ] );
# evolve 1000 generations
$ga->evolve(1);
my $count = 0;
for($ga->getFittest(4)){
$count++ if $ga->as_value($_) == $Win;
}
ok($count >= 4);
t/15_bitvectors_constant_length_MCE.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/15_bitvectors_constant_length_MCE.t view on Meta::CPAN
[ qw( 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/16_bitvectors_constant_length_-_native_arrays.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/16_bitvectors_constant_length_-_native_arrays.t view on Meta::CPAN
[ qw( 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
t/17_bitvectors_constant_length_MCE_-_native_arrays.t view on Meta::CPAN
return $counter;
}
sub fitness {
my ($ga, $chromosome) = @_;
return sum(scalar $ga->as_array($chromosome));
}
sub terminate {
my ($ga) = @_;
return 1 if $Win == $ga->as_value($ga->getFittest);
return;
}
my $ga = AI::Genetic::Pro->new(
-fitness => \&fitness, # fitness function
-terminate => \&terminate, # terminate function
-type => 'bitvector', # type of chromosomes
-population => 100, # population
-crossover => 0.9, # probab. of crossover
-mutation => 0.05, # probab. of mutation
t/17_bitvectors_constant_length_MCE_-_native_arrays.t view on Meta::CPAN
[ qw( 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 ) ],
[ qw( 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 ) ],
);
$ga->inject(\@helper);
# evolve 1000 generations
$ga->evolve(1000);
ok($Win == $ga->as_value($ga->getFittest));
( run in 0.387 second using v1.01-cache-2.11-cpan-3cd7ad12f66 )