Algorithm-DecisionTree

 view release on metacpan or  search on metacpan

META.json  view on Meta::CPAN

         }
      },
      "configure" : {
         "requires" : {
            "ExtUtils::MakeMaker" : "0"
         }
      },
      "runtime" : {
         "requires" : {
            "List::Util" : "1.33",
            "Math::Random" : "0"
         }
      }
   },
   "release_status" : "stable",
   "version" : "3.43",
   "x_serialization_backend" : "JSON::PP version 2.27300"
}

META.yml  view on Meta::CPAN

meta-spec:
  url: http://module-build.sourceforge.net/META-spec-v1.4.html
  version: '1.4'
name: Algorithm-DecisionTree
no_index:
  directory:
    - t
    - inc
requires:
  List::Util: '1.33'
  Math::Random: '0'
version: '3.43'
x_serialization_backend: 'CPAN::Meta::YAML version 0.012'

Makefile.PL  view on Meta::CPAN

#if ($^V lt v5.10) {
#   die("Algorithm::DecisionTree has been tested on Perl 5.10.1.\n" .
#   "Your perl version is $].\n");
#}

copy("perl/MANIFEST.perl","MANIFEST");

WriteMakefile(
    NAME         => 'Algorithm::DecisionTree',
    VERSION_FROM => 'lib/Algorithm/DecisionTree.pm',
    PREREQ_PM    => { Math::Random => 0, 
                      List::Util   => 1.33,
                    },
    AUTHOR       => 'Avinash Kak (kak@purdue.edu)',
    ABSTRACT     => 'A Perl module for decision-tree based classification of multidimensional data',
    clean        => {FILES => join(" ",
                                   map { "$_ */$_ */*/$_" }
                                   qw( *% *.html *.b[ac]k *.old *.orig ) )
                    },
);

lib/Algorithm/DecisionTree.pm  view on Meta::CPAN

        }
    }
    $self->{_class_names}        =   \@class_names;
    $self->{_class_names_and_priors}   = \%class_names_and_priors;
    $self->{_features_with_value_range}   = \%features_with_value_range;
    $self->{_classes_and_their_param_values} = \%classes_and_their_param_values;
    $self->{_features_ordered} = \@features_ordered;
}

##  After the parameter file is parsed by the previous method, this method calls on
##  Math::Random::random_multivariate_normal() to generate the training and test data
##  samples. Your training and test data can be of any number of of dimensions, can
##  have any mean, and any covariance.  The training and test data must obviously be
##  drawn from the same distribution.
sub gen_numeric_training_and_test_data_and_write_to_csv {
    use Math::Random;
    my $self = shift;
    my %training_samples_for_class;
    my %test_samples_for_class;
    foreach my $class_name (@{$self->{_class_names}}) {
        $training_samples_for_class{$class_name} = [];
        $test_samples_for_class{$class_name} = [];
    }
    foreach my $class_name (keys %{$self->{_classes_and_their_param_values}}) {
        my @mean = @{$self->{_classes_and_their_param_values}->{$class_name}->{'mean'}};
        my @covariance = @{$self->{_classes_and_their_param_values}->{$class_name}->{'covariance'}};
        my @new_training_data = Math::Random::random_multivariate_normal(
              $self->{_number_of_samples_for_training} * $self->{_class_names_and_priors}->{$class_name},
              @mean, @covariance );
        my @new_test_data = Math::Random::random_multivariate_normal(
              $self->{_number_of_samples_for_testing} * $self->{_class_names_and_priors}->{$class_name},
              @mean, @covariance );
        if ($self->{_debug}) {
            print "training data for class $class_name:\n";
            foreach my $x (@new_training_data) {print "@$x\n";}
            print "\n\ntest data for class $class_name:\n";
            foreach my $x (@new_test_data) {print "@$x\n";}
        }
        $training_samples_for_class{$class_name} = \@new_training_data;
        $test_samples_for_class{$class_name} = \@new_test_data;

 view all matches for this distribution
 view release on metacpan -  search on metacpan