view release on metacpan or  search on metacpan

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

package Algorithm::ExpectationMaximization;

#---------------------------------------------------------------------------
# Copyright (c) 2014 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::ExpectationMaximization is a pure Perl implementation for
# Expectation-Maximization based clustering of multi-dimensional data that can be
# modeled as a Gaussian mixture.
# ---------------------------------------------------------------------------

use 5.10.0;
use strict;
use warnings;
use Carp;
use File::Basename;
use Math::Random;
use Graphics::GnuplotIF;
use Math::GSL::Matrix;
use Scalar::Util 'blessed';

our $VERSION = '1.22';

# from perl docs:
my $_num_regex =  '^[+-]?\ *(\d+(\.\d*)?|\.\d+)([eE][+-]?\d+)?$'; 

# Constructor:
sub new { 
    my ($class, %args) = @_;
    my @params = keys %args;
    croak "\nYou have used a wrong name for a keyword argument " .
          "--- perhaps a misspelling\n" 
          if check_for_illegal_params(@params) == 0;
    bless {
        _datafile         =>   $args{datafile} || croak("datafile required"),
        _mask             =>   $args{mask}     || croak("mask required"),
        _K                =>   $args{K}       || croak("number of clusters required"),
        _terminal_output  =>   $args{terminal_output}   || 0,
        _seeding          =>   $args{seeding}           || 'random',
        _seed_tags        =>   $args{seed_tags}         || [],
        _max_em_iterations=>   $args{max_em_iterations} || 100,
        _class_priors     =>   $args{class_priors}      || [],
        _debug            =>   $args{debug}             || 0,
        _N                =>   0,
        _data             =>   {},
        _data_id_tags     =>   [],
        _clusters         =>   [],
        _cluster_centers  =>   [],
        _data_dimensions  =>   0,
        _cluster_normalizers            => [],
        _cluster_means                  => [],
        _cluster_covariances            => [],
        _class_labels_for_data          => {},
        _class_probs_at_each_data_point => {},
        _expected_class_probs           => {}, 
        _old_priors                     => [],
        _old_old_priors                 => [],
        _fisher_quality_vs_iteration    => [],
        _mdl_quality_vs_iterations      => [],
    }, $class;
}


sub read_data_from_file {
    my $self = shift;