A greedy averaged perceptron tagger. More...

#include <perceptron.h>

Classes
struct	training_options
	Training options required for learning a tagger. More...

Public Member Functions
	perceptron ()
	Default constructor.

	perceptron (const std::string &prefix)
	Loads a perceptron tagger from a given prefix. More...

void	tag (sequence &seq) const
	Tags a sequence. More...

void	train (std::vector< sequence > &sequences, const training_options &options)
	Trains the tagger on a set of sequences using the given options. More...

void	save (const std::string &prefix) const
	Saves the model to the folder specified by prefix. More...

Private Attributes
sequence_analyzer	analyzer_
	The analyzer used for feature generation.

classify::linear_model< feature_id, double, label_id >	model_
	The model storage.

Detailed Description

A greedy averaged perceptron tagger.

Constructor & Destructor Documentation

meta::sequence::perceptron::perceptron ( const std::string & prefix )

Loads a perceptron tagger from a given prefix.

Parameters

prefix The folder that contains the tagger model

void meta::sequence::perceptron::tag ( sequence & seq ) const

Tags a sequence.

This sets both the label and tag of the sequence's observations.

Parameters

seq	The sequence to be tagged

void meta::sequence::perceptron::train	(	std::vector< sequence > &	sequences,
		const training_options &	options
	)

Trains the tagger on a set of sequences using the given options.

The sequences given for training will be analyzed by the tagger internally, so they do not need to be analyzed ahead of time.

Parameters

sequences	The training data
options	THe training options

void meta::sequence::perceptron::save ( const std::string & prefix ) const

Saves the model to the folder specified by prefix.

Both the tagger and its analyzer are serialized.

Parameters

prefix The folder to save the model to

The documentation for this class was generated from the following files: