view release on metacpan or  search on metacpan

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

Depending on the number of instances (and to a lesser extent the total
number of attributes), this method might take a while.  If you want to
train the model only once and save it for later re-use in a different
context, see the C<write_model()> and C<read_model()> methods.

=item is_trained()

Returns a boolean value indicating whether or not C<train()> has been
called on this model.

=item predict(attributes => \%y)

After C<train()> has been called, the model may be applied to
previously-unseen combinations of attributes.  The C<predict()> method
accepts an C<attributes> parameter just like C<add_instance()>, and
returns its best prediction of the label that would apply to the given
attributes.  The sign of the returned label (positive or negative)
indicates whether the new instance is considered a positive or
negative instance, and the magnitude of the label corresponds in some
way to the confidence with which the model is making that assertion.

=item predict_i(\@indices, \@values)

This is just like C<predict()>, but bypasses all the string-to-integer
mapping of feature names.  See also C<add_instance_i()>.

=item write_model($file)

Saves the given trained model to the file C<$file>.  The model may
later be re-loaded using the C<read_model()> method.  The model is
written using SVMLight's C<write_model()> C function, so it will be
fully compatible with SVMLight command-line tools like
C<svm_classify>.

=item read_model($file)

Reads a model that has previously been written with C<write_model()>:

  my $m = Algorithm::SVMLight->new();
  $m->read_model($file);

The model file is read using SVMLight's C<read_model()> C function, so
if you want to, you could initially create the model with one of
SVMLight's command-line tools like C<svm_learn>.

=item get_linear_weights()

After training a linear model (or reading in a model file), this
method will return a reference to an array containing the linear
weights of the model.  This can be useful for model inspection, to see
which features are having the greatest impact on decision-making.

 my $arrayref = $m->get_linear_weights();

The first element (position 0) of the array will be the threshold
C<b>, and the rest of the elements will be the weights themselves.
Thus from 1 upward, the indices align with SVMLight's internal
indices.

If the model has not yet been trained, or if the kernel type is not
linear, an exception will be thrown.

=item feature_names()

Returns a list of feature names that have been fed to
C<add_instance()> as keys of the C<attribute> parameter, or in a
scalar context the number of such names.

=item num_features()

Returns the number of features known to this model.  Note that if you
use C<add_instance_i()> or C<read_instances()>, some of the features
may never actually have been I<seen> before, because you could add
instances with only indices 2, 5, and 37, never having added any
instances with the indices in between, but C<num_features()> will
return 37 in this case.  This is because after training, an instance
could be passed to the C<predict()> method with real values for these
previously unseen features.  If you just use C<add_instance()>
instead, you'll probably never run into this issue, and in a scalar
context C<num_features()> will look just like C<feature_names()>.

=item num_instances()

Returns the number of training instances known to the model.  It
should be fine to call this method either before or after training
actually occurs.

=back

=head1 SEE ALSO

L<Algorithm::NaiveBayes>, L<AI::DecisionTree>

L<http://svmlight.joachims.org/>

=head1 AUTHOR

Ken Williams, E<lt>kwilliams@cpan.orgE<gt>

=head1 COPYRIGHT AND LICENSE

The C<Algorithm::SVMLight> perl interface is copyright (C) 2005-2008
Thomson Legal & Regulatory, and written by Ken Williams.  It is free
software; you can redistribute it and/or modify it under the same
terms as C<perl> itself.

Thorsten Joachims and/or Cornell University of Ithaca, NY control the
copyright of SVMLight itself - you will find full copyright and
license information in its distribution.  You are responsible for
obtaining an appropriate license for SVMLight if you intend to use
C<Algorithm::SVMLight>.  In particular, please note that SVMLight "is
granted free of charge for research and education purposes. However
you must obtain a license from the author to use it for commercial
purposes."

To avoid any copyright clashes, the F<SVMLight.patch> file distributed
here is granted under the same license terms as SVMLight itself.

=cut