Algorithm-LinearManifoldDataClusterer

 view release on metacpan or  search on metacpan

examples/example1.pl  view on Meta::CPAN

#!/usr/bin/perl -w

#use lib '../blib/lib', '../blib/arch';


##  example1.pl

##  Highlights:
##
##    ---  The data file contains 498 samples in three small clusters 
##         on the surface of a sphere
##
##    ---  Note the use of 0.001 for delta_reconstruction_error

examples/example2.pl  view on Meta::CPAN

#!/usr/bin/perl -w

#use lib '../blib/lib', '../blib/arch';

##  example2.pl

##  Highlights:
##
##    ---  The data file contains 3000 samples in three large
##         clusters on the surface of a sphere
##
##    ---  Note the use of 0.012 for delta_reconstruction_error

examples/example3.pl  view on Meta::CPAN

#!/usr/bin/perl -w

#use lib '../blib/lib', '../blib/arch';

##  example3.pl

##  Highlights:
##
##    --- The data file contains 1000 samples in four small
##        clusters on the surface of a sphere
##
##    ---  Note the use of 0.002 for delta_reconstruction_error

examples/example4.pl  view on Meta::CPAN

#!/usr/bin/perl -w

#use lib '../blib/lib', '../blib/arch';


##  example1.pl

##  Highlights:
##
##    ---  The main highlight here is the use of the auto_retry_clusterer()
##         method for automatically invoking the clusterer repeatedly 
##         should it fail on account of the Fail-First bias built into
##         the code.

examples/generate_data_on_a_sphere.pl  view on Meta::CPAN

#!/usr/bin/perl -w

## generate_data_on_a_sphere.pl

use lib '../blib/lib', '../blib/arch';


##    The purpose of this script is to generate multivariate Gaussian data
##    on a spherical surface and, subsequently, to also visualize this
##    data.  Read the comment block attached to the subroutine
##    `gen_data_and_write_to_csv() in the main module file.  That
##    subroutine randomly chooses a number of directions equal to the value
##    of the number_of_clusters_on_sphere.  It also put together 2x2
##    covariance matrices for each of these clusters.  Subsquently, the
##    Random module is called to yield multivariates samples for each

t/test.t  view on Meta::CPAN

use Test::Simple tests => 3;

use lib '../blib/lib','../blib/arch';

use Algorithm::LinearManifoldDataClusterer;

# Test 1 (Data Generation):

my $datafile = "__datadump.csv";
#my $datagen = Algorithm::LinearManifoldDataClusterer::DataGenerator->new( 
my $datagen = DataGenerator->new( 
                                  output_file => $datafile,
                                  total_number_of_samples_needed => 100,