view release on metacpan or  search on metacpan

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

        print "\nAll class names: @{$self->{_all_trees}->{0}->{_class_names}}\n";
        print "\nEach sample data record:\n";
        foreach my $kee (sort {sample_index($a) <=> sample_index($b)} keys %{$self->{_all_trees}->{0}->{_training_data_hash}}) {
            print "$kee    =>   @{$self->{_all_trees}->{0}->{_training_data_hash}->{$kee}}\n";
        }
        print "\nclass label for each data sample:\n";        
        foreach my $kee (sort {sample_index($a) <=> sample_index($b)} keys %{$self->{_all_trees}->{0}->{_samples_class_label_hash}}) {
            print "$kee    =>   $self->{_all_trees}->{0}->{_samples_class_label_hash}->{$kee}\n";            
        }
        print "\nfeatures and the values taken by them:\n";
        for my $kee  (sort keys %{$self->{_all_trees}->{0}->{_features_and_values_hash}}) {
            print "$kee    =>   @{$self->{_all_trees}->{0}->{_features_and_values_hash}->{$kee}}\n";                        
        }
        print "\nnumeric features and their ranges:\n";
        for my $kee  (sort keys %{$self->{_all_trees}->{0}->{_numeric_features_valuerange_hash}}) {
            print "$kee    =>   @{$self->{_all_trees}->{0}->{_numeric_features_valuerange_hash}->{$kee}}\n";
        }
        print "\nunique values for the features:\n";
        for my $kee  (sort keys %{$self->{_all_trees}->{0}->{_features_and_unique_values_hash}}) {
            print "$kee    =>   @{$self->{_all_trees}->{0}->{_features_and_unique_values_hash}->{$kee}}\n";  
        }
        print "\nnumber of unique values in each feature:\n";        
        for my $kee  (sort keys %{$self->{_all_trees}->{0}->{_feature_values_how_many_uniques_hash}}) {
            print "$kee    =>   $self->{_all_trees}->{0}->{_feature_values_how_many_uniques_hash}->{$kee}\n";
        }
    }
}

sub show_training_data_for_base_tree {
    my $self = shift;
    $self->{_all_trees}->{0}->show_training_data();
}

sub calculate_first_order_probabilities_and_class_priors {
    my $self = shift;
    $self->{_all_trees}->{0}->calculate_first_order_probabilities();
    $self->{_all_trees}->{0}->calculate_class_priors();
    $self->{_sample_selection_probs}->{0} =  {map { $_ => 1.0/@{$self->{_all_sample_names}} } @{$self->{_all_sample_names}}};
}

sub construct_base_decision_tree {
    my $self = shift;
    $self->{_root_nodes}->{0} = $self->{_all_trees}->{0}->construct_decision_tree_classifier();
}

sub display_base_decision_tree {
    my $self = shift;
    $self->{_root_nodes}->{0}->display_decision_tree("     ");
}

sub construct_cascade_of_trees {
    my $self = shift;
    $self->{_training_samples}->{0} = $self->{_all_sample_names};
    $self->{_misclassified_samples}->{0} = $self->evaluate_one_stage_of_cascade($self->{_all_trees}->{0}, $self->{_root_nodes}->{0});
    if ($self->{_stagedebug}) {
        $self->show_class_labels_for_misclassified_samples_in_stage(0);
        print "\n\nSamples misclassified by base classifier: @{$self->{_misclassified_samples}->{0}}\n";
        my $how_many = @{$self->{_misclassified_samples}->{0}};
        print "\nNumber of misclassified samples: $how_many\n";
    }
    my $misclassification_error_rate = reduce {$a+$b} map {$self->{_sample_selection_probs}->{0}->{$_}} @{$self->{_misclassified_samples}->{0}};
    print "\nMisclassification_error_rate for base classifier: $misclassification_error_rate\n" if $self->{_stagedebug};
    $self->{_trust_factors}->{0} = 0.5 * log((1-$misclassification_error_rate)/$misclassification_error_rate);
    print "\nBase class trust factor: $self->{_trust_factors}->{0}\n"  if $self->{_stagedebug};
    foreach my $stage_index (1 .. $self->{_how_many_stages} - 1) {
        print "\n\n========================== Constructing stage indexed $stage_index =========================\n"
              if $self->{_stagedebug};
        $self->{_sample_selection_probs}->{$stage_index} =  { map {$_ =>  $self->{_sample_selection_probs}->{$stage_index-1}->{$_} *   exp(-1.0 * $self->{_trust_factors}->{$stage_index - 1} *  (contained_in($_, @{$self->{_misclassified_samples}->{$st...
        my $normalizer = reduce {$a + $b} values %{$self->{_sample_selection_probs}->{$stage_index}};
        print "\nThe normalizer is: $normalizer\n"  if $self->{_stagedebug};
        map {$self->{_sample_selection_probs}->{$stage_index}->{$_}  /= $normalizer} keys %{$self->{_sample_selection_probs}->{$stage_index}};
        my @training_samples_this_stage = ();
        my $sum_of_probs = 0.0;
        foreach my $sample (sort {$self->{_sample_selection_probs}->{$stage_index}->{$b} <=> $self->{_sample_selection_probs}->{$stage_index}->{$a}} keys %{$self->{_sample_selection_probs}->{$stage_index}}) {
            $sum_of_probs += $self->{_sample_selection_probs}->{$stage_index}->{$sample};
            push @training_samples_this_stage, $sample if $sum_of_probs < 0.5;
            last if $sum_of_probs > 0.5;
        }
        $self->{_training_samples}->{$stage_index} = [sort {sample_index($a) <=> sample_index($b)} @training_samples_this_stage];
        if ($self->{_stagedebug}) {
            print "\nTraining samples for stage $stage_index: @{$self->{_training_samples}->{$stage_index}}\n\n";
            my $num_of_training_samples = @{$self->{_training_samples}->{$stage_index}};
            print "\nNumber of training samples this stage $num_of_training_samples\n\n";
        }
        # find intersection of two sets:
        my %misclassified_samples = map {$_ => 1} @{$self->{_misclassified_samples}->{$stage_index-1}};
        my @training_samples_selection_check = grep $misclassified_samples{$_}, @{$self->{_training_samples}->{$stage_index}};
        if ($self->{_stagedebug}) {
            my @training_in_misclassified = sort {sample_index($a) <=> sample_index($b)} @training_samples_selection_check;
            print "\nTraining samples in the misclassified set: @training_in_misclassified\n";
            my $how_many = @training_samples_selection_check;
            print "\nNumber_of_miscalssified_samples_in_training_set: $how_many\n";
        }
        my $dt_this_stage = Algorithm::DecisionTree->new('boostingmode');
        $dt_this_stage->{_training_data_hash} = { map {$_ => $self->{_all_training_data}->{$_} } @{$self->{_training_samples}->{$stage_index}} };

        $dt_this_stage->{_class_names} = $self->{_all_trees}->{0}->{_class_names};
        $dt_this_stage->{_feature_names} = $self->{_all_trees}->{0}->{_feature_names};
        $dt_this_stage->{_entropy_threshold} = $self->{_all_trees}->{0}->{_entropy_threshold};
        $dt_this_stage->{_max_depth_desired} = $self->{_all_trees}->{0}->{_max_depth_desired};        
        $dt_this_stage->{_symbolic_to_numeric_cardinality_threshold} = $self->{_all_trees}->{0}->{_symbolic_to_numeric_cardinality_threshold};
        $dt_this_stage->{_samples_class_label_hash} = {map {$_ => $self->{_all_trees}->{0}->{_samples_class_label_hash}->{$_}} keys %{$dt_this_stage->{_training_data_hash}}};
        $dt_this_stage->{_features_and_values_hash} = {map {$_ => []} keys %{$self->{_all_trees}->{0}->{_features_and_values_hash}}};
        my $pattern = '(\S+)\s*=\s*(\S+)';        
        foreach my $sample (sort {sample_index($a) <=> sample_index($b)} keys %{$dt_this_stage->{_training_data_hash}}) { 
            foreach my $feature_and_value (@{$dt_this_stage->{_training_data_hash}->{$sample}}) {
                $feature_and_value =~ /$pattern/;
                my ($feature, $value) = ($1, $2);
                push @{$dt_this_stage->{_features_and_values_hash}->{$feature}}, $value if $value ne 'NA';
            }
        }
        $dt_this_stage->{_features_and_unique_values_hash} = {map {my $feature = $_; $feature => [sort keys %{{map {$_ => 1} @{$dt_this_stage->{_features_and_values_hash}->{$feature}}}}]} keys %{$dt_this_stage->{_features_and_values_hash}}};
        $dt_this_stage->{_numeric_features_valuerange_hash} = {map {$_ => []} keys %{$self->{_all_trees}->{0}->{_numeric_features_valuerange_hash}}};
        $dt_this_stage->{_numeric_features_valuerange_hash} = {map {my $feature = $_; $feature =>  [min(@{$dt_this_stage->{_features_and_unique_values_hash}->{$feature}}), max(@{$dt_this_stage->{_features_and_unique_values_hash}->{$feature}})]} keys ...
        if ($self->{_stagedebug}) {
            print "\n\nPrinting features and their values in the training set:\n\n";
            foreach my $kee (sort keys %{$dt_this_stage->{_features_and_values_hash}}) {
                print "$kee   =>  @{$dt_this_stage->{_features_and_values_hash}->{$kee}}\n";
            }
            print "\n\nPrinting unique values for features:\n\n";
            foreach my $kee (sort keys %{$dt_this_stage->{_features_and_unique_values_hash}}) {
                print "$kee   =>  @{$dt_this_stage->{_features_and_unique_values_hash}->{$kee}}\n";            
            }
            print "\n\nPrinting unique value ranges for features:\n\n";
            foreach my $kee (sort keys %{$dt_this_stage->{_numeric_features_valuerange_hash}}) {
                print "$kee   =>  @{$dt_this_stage->{_numeric_features_valuerange_hash}->{$kee}}\n";            
            }
        }
        $dt_this_stage->{_feature_values_how_many_uniques_hash} = {map {$_ => undef} keys %{$self->{_all_trees}->{0}->{_features_and_unique_values_hash}}};
        $dt_this_stage->{_feature_values_how_many_uniques_hash} = {map {$_ => scalar @{$dt_this_stage->{_features_and_unique_values_hash}->{$_}}} keys %{$self->{_all_trees}->{0}->{_features_and_unique_values_hash}}};
        $dt_this_stage->calculate_first_order_probabilities();
        $dt_this_stage->calculate_class_priors();
        print "\n\n>>>>>>>Done with the initialization of the tree for stage $stage_index<<<<<<<<<<\n" if $self->{_stagedebug};
        my $root_node_this_stage = $dt_this_stage->construct_decision_tree_classifier();
        $root_node_this_stage->display_decision_tree("     ") if $self->{_stagedebug};

        $self->{_all_trees}->{$stage_index} = $dt_this_stage;
        $self->{_root_nodes}->{$stage_index} = $root_node_this_stage;
        $self->{_misclassified_samples}->{$stage_index} = $self->evaluate_one_stage_of_cascade($self->{_all_trees}->{$stage_index}, $self->{_root_nodes}->{$stage_index});
        if ($self->{_stagedebug}) {
            print "\nSamples misclassified by stage $stage_index classifier: @{$self->{_misclassified_samples}->{$stage_index}}\n";
            printf("\nNumber of misclassified samples: %d\n", scalar @{$self->{_misclassified_samples}->{$stage_index}});
            $self->show_class_labels_for_misclassified_samples_in_stage($stage_index);
        }
        my $misclassification_error_rate = reduce {$a+$b} map {$self->{_sample_selection_probs}->{$stage_index}->{$_}} @{$self->{_misclassified_samples}->{$stage_index}};
        print "\nStage $stage_index misclassification_error_rate: $misclassification_error_rate\n" if $self->{_stagedebug};

        $self->{_trust_factors}->{$stage_index} = 0.5 * log((1-$misclassification_error_rate)/$misclassification_error_rate);
        print "\nStage $stage_index trust factor: $self->{_trust_factors}->{$stage_index}\n"  if $self->{_stagedebug};
    }
}

sub evaluate_one_stage_of_cascade {
    my $self = shift;
    my $trainingDT = shift;
    my $root_node = shift;
    my @misclassified_samples = ();
    foreach my $test_sample_name (@{$self->{_all_sample_names}}) {
        my @test_sample_data = @{$self->{_all_trees}->{0}->{_training_data_hash}->{$test_sample_name}};
        print "original data in $test_sample_name:@test_sample_data\n" if $self->{_stagedebug};
        @test_sample_data = map {$_ if $_ !~ /=NA$/} @test_sample_data;
        print "$test_sample_name: @test_sample_data\n" if $self->{_stagedebug}; 
        my %classification = %{$trainingDT->classify($root_node, \@test_sample_data)};
        my @solution_path = @{$classification{'solution_path'}};                                  
        delete $classification{'solution_path'};                                              
        my @which_classes = keys %classification;
        @which_classes = sort {$classification{$b} <=> $classification{$a}} @which_classes;
        my $most_likely_class_label = $which_classes[0];
        if ($self->{_stagedebug}) {
            print "\nClassification:\n\n";
            print "     class                         probability\n";
            print "     ----------                    -----------\n";
            foreach my $which_class (@which_classes) {
                my $classstring = sprintf("%-30s", $which_class);
                my $valuestring = sprintf("%-30s", $classification{$which_class});
                print "     $classstring $valuestring\n";
            }
            print "\nSolution path in the decision tree: @solution_path\n";
            print "\nNumber of nodes created: " . $root_node->how_many_nodes() . "\n";
        }
        my $true_class_label_for_test_sample = $self->{_all_trees}->{0}->{_samples_class_label_hash}->{$test_sample_name};
        printf("%s:   true_class: %s    estimated_class: %s\n", $test_sample_name, $true_class_label_for_test_sample, $most_likely_class_label) if $self->{_stagedebug};
        push @misclassified_samples, $test_sample_name if $true_class_label_for_test_sample ne $most_likely_class_label;
    }
    return [sort {sample_index($a) <=> sample_index($b)} @misclassified_samples];
}

sub show_class_labels_for_misclassified_samples_in_stage {
    my $self = shift;
    my $stage_index = shift;
    die "\nYou must first call 'construct_cascade_of_trees()' before invoking 'show_class_labels_for_misclassified_samples_in_stage()'" unless @{$self->{_misclassified_samples}->{0}} > 0;
    my @classes_for_misclassified_samples = ();
    my @just_class_labels = ();

    for my $sample (@{$self->{_misclassified_samples}->{$stage_index}}) {    
        my $true_class_label_for_sample = $self->{_all_trees}->{0}->{_samples_class_label_hash}->{$sample};            
        push @classes_for_misclassified_samples, sprintf("%s => %s", $sample, $true_class_label_for_sample);
        push @just_class_labels, $true_class_label_for_sample; 
    }
    print "\nSamples misclassified by the classifier for Stage $stage_index: @{$self->{_misclassified_samples}->{$stage_index}}\n";
    my $how_many = @{$self->{_misclassified_samples}->{$stage_index}};
    print "\nNumber of misclassified samples: $how_many\n";
    print "\nShowing class labels for samples misclassified by stage $stage_index: ";
    print "\nClass labels for samples: @classes_for_misclassified_samples\n";
    my @class_names_unique =  sort keys %{{map {$_ => 1} @just_class_labels}};
    print "\nClass names (unique) for misclassified samples: @class_names_unique\n";
    print "\nFinished displaying class labels for samples misclassified by stage $stage_index\n\n";
}