view release on metacpan or  search on metacpan

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

package Algorithm::DecisionTree;

#--------------------------------------------------------------------------------------
# Copyright (c) 2017 Avinash Kak. All rights reserved.  This program is free
# software.  You may modify and/or distribute it under the same terms as Perl itself.
# This copyright notice must remain attached to the file.
#
# Algorithm::DecisionTree is a Perl module for decision-tree based classification of 
# multidimensional data.
# -------------------------------------------------------------------------------------

#use 5.10.0;
use strict;
use warnings;
use Carp;

our $VERSION = '3.43';

############################################   Constructor  ##############################################
sub new { 
    my ($class, %args, $eval_or_boosting_mode);
    if (@_ % 2 != 0) {
        ($class, %args) = @_;
    } else {
        $class = shift;
        $eval_or_boosting_mode = shift;
        die unless $eval_or_boosting_mode eq 'evalmode' || $eval_or_boosting_mode eq 'boostingmode';
        die "Only one string arg allowed in eval and boosting modes" if @_;
    }
    unless ($eval_or_boosting_mode) {
        my @params = keys %args;
        croak "\nYou have used a wrong name for a keyword argument --- perhaps a misspelling\n" 
                           if check_for_illegal_params2(@params) == 0;
    }
    bless {
        _training_datafile                   =>    $args{training_datafile}, 
        _entropy_threshold                   =>    $args{entropy_threshold} || 0.01,
        _max_depth_desired                   =>    exists $args{max_depth_desired} ? 
                                                                       $args{max_depth_desired} : undef,
        _debug1                              =>    $args{debug1} || 0,
        _debug2                              =>    $args{debug2} || 0,
        _debug3                              =>    $args{debug3} || 0,
        _csv_class_column_index              =>    $args{csv_class_column_index} || undef,
        _csv_columns_for_features            =>    $args{csv_columns_for_features} || undef,
        _symbolic_to_numeric_cardinality_threshold
                                             =>    $args{symbolic_to_numeric_cardinality_threshold} || 10,
        _number_of_histogram_bins            =>    $args{number_of_histogram_bins} || undef,
        _csv_cleanup_needed                  =>    $args{csv_cleanup_needed} || 0,
        _training_data                       =>    [],
        _root_node                           =>    undef,
        _probability_cache                   =>    {},
        _entropy_cache                       =>    {},
        _training_data_hash                  =>    {},
        _features_and_values_hash            =>    {},
        _samples_class_label_hash            =>    {},
        _class_names                         =>    [],
        _class_priors                        =>    [],
        _class_priors_hash                   =>    {},
        _feature_names                       =>    [],
        _numeric_features_valuerange_hash    =>    {},
        _sampling_points_for_numeric_feature_hash      =>      {},
        _feature_values_how_many_uniques_hash          =>      {},
        _prob_distribution_numeric_features_hash       =>      {},
        _histogram_delta_hash                          =>      {},
        _num_of_histogram_bins_hash                    =>      {},
    }, $class;