opennlp.tools.ml.maxent

Class GIS

java.lang.Object

opennlp.tools.ml.AbstractTrainer

opennlp.tools.ml.AbstractEventTrainer

opennlp.tools.ml.maxent.GIS

All Implemented Interfaces:

EventTrainer

public class GIS
extends AbstractEventTrainer

A Factory class which uses instances of GISTrainer to create and train GISModels.

Field Summary

Fields

Modifier and Type	Field and Description
static String	MAXENT_VALUE
static boolean	PRINT_MESSAGES Set this to false if you don't want messages about the progress of model training displayed.
static double	SMOOTHING_OBSERVATION If we are using smoothing, this is used as the "number" of times we want the trainer to imagine that it saw a feature that it actually didn't see.

Fields inherited from class opennlp.tools.ml.AbstractEventTrainer

DATA_INDEXER_ONE_PASS_VALUE, DATA_INDEXER_PARAM, DATA_INDEXER_TWO_PASS_VALUE

Fields inherited from class opennlp.tools.ml.AbstractTrainer

ALGORITHM_PARAM, CUTOFF_DEFAULT, CUTOFF_PARAM, ITERATIONS_DEFAULT, ITERATIONS_PARAM, TRAINER_TYPE_PARAM

Fields inherited from interface opennlp.tools.ml.EventTrainer

EVENT_VALUE

Constructor Summary

Constructors

Constructor and Description
GIS()

Method Summary

Methods

Modifier and Type	Method and Description
AbstractModel	doTrain(DataIndexer indexer)
boolean	isSortAndMerge()
boolean	isValid()
static GISModel	trainModel(int iterations, DataIndexer indexer) Train a model using the GIS algorithm.
static GISModel	trainModel(int iterations, DataIndexer indexer, boolean smoothing) Train a model using the GIS algorithm.
static GISModel	trainModel(int iterations, DataIndexer indexer, boolean printMessagesWhileTraining, boolean smoothing, Prior modelPrior, int cutoff) Train a model using the GIS algorithm.
static GISModel	trainModel(int iterations, DataIndexer indexer, boolean printMessagesWhileTraining, boolean smoothing, Prior modelPrior, int cutoff, int threads) Train a model using the GIS algorithm.
static GISModel	trainModel(int iterations, DataIndexer indexer, Prior modelPrior, int cutoff) Train a model using the GIS algorithm with the specified number of iterations, data indexer, and prior.
static GISModel	trainModel(ObjectStream<Event> eventStream) Train a model using the GIS algorithm, assuming 100 iterations and no cutoff.
static GISModel	trainModel(ObjectStream<Event> eventStream, boolean smoothing) Train a model using the GIS algorithm, assuming 100 iterations and no cutoff.
static GISModel	trainModel(ObjectStream<Event> eventStream, int iterations, int cutoff) Train a model using the GIS algorithm.
static GISModel	trainModel(ObjectStream<Event> eventStream, int iterations, int cutoff, boolean smoothing, boolean printMessagesWhileTraining) Train a model using the GIS algorithm.
static GISModel	trainModel(ObjectStream<Event> eventStream, int iterations, int cutoff, double sigma) Train a model using the GIS algorithm.

Methods inherited from class opennlp.tools.ml.AbstractEventTrainer

getDataIndexer, train

Methods inherited from class opennlp.tools.ml.AbstractTrainer

getAlgorithm, getCutoff, getIterations, init

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface opennlp.tools.ml.EventTrainer

init

Field Detail

MAXENT_VALUE

public static final String MAXENT_VALUE

See Also:

Constant Field Values

PRINT_MESSAGES

public static boolean PRINT_MESSAGES

Set this to false if you don't want messages about the progress of model training displayed. Alternately, you can use the overloaded version of trainModel() to conditionally enable progress messages.

SMOOTHING_OBSERVATION

public static double SMOOTHING_OBSERVATION

If we are using smoothing, this is used as the "number" of times we want the trainer to imagine that it saw a feature that it actually didn't see. Defaulted to 0.1.

Constructor Detail

GIS

public GIS()

Method Detail

isValid

public boolean isValid()

Overrides:

isValid in class AbstractEventTrainer

isSortAndMerge

public boolean isSortAndMerge()

Specified by:

isSortAndMerge in class AbstractEventTrainer

doTrain

public AbstractModel doTrain(DataIndexer indexer)
                      throws IOException

Specified by:

doTrain in class AbstractEventTrainer

Throws:

IOException

trainModel

public static GISModel trainModel(ObjectStream<Event> eventStream)
                           throws IOException

Train a model using the GIS algorithm, assuming 100 iterations and no cutoff.

Parameters:

eventStream - The EventStream holding the data on which this model will be trained.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

Throws:

IOException

trainModel

public static GISModel trainModel(ObjectStream<Event> eventStream,
                  boolean smoothing)
                           throws IOException

Train a model using the GIS algorithm, assuming 100 iterations and no cutoff.

Parameters:

eventStream - The EventStream holding the data on which this model will be trained.

smoothing - Defines whether the created trainer will use smoothing while training the model.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

Throws:

IOException

trainModel

public static GISModel trainModel(ObjectStream<Event> eventStream,
                  int iterations,
                  int cutoff)
                           throws IOException

Train a model using the GIS algorithm.

Parameters:

eventStream - The EventStream holding the data on which this model will be trained.

iterations - The number of GIS iterations to perform.

cutoff - The number of times a feature must be seen in order to be relevant for training.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

Throws:

IOException

trainModel

public static GISModel trainModel(ObjectStream<Event> eventStream,
                  int iterations,
                  int cutoff,
                  boolean smoothing,
                  boolean printMessagesWhileTraining)
                           throws IOException

Train a model using the GIS algorithm.

Parameters:

eventStream - The EventStream holding the data on which this model will be trained.

iterations - The number of GIS iterations to perform.

cutoff - The number of times a feature must be seen in order to be relevant for training.

smoothing - Defines whether the created trainer will use smoothing while training the model.

printMessagesWhileTraining - Determines whether training status messages are written to STDOUT.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

Throws:

IOException

trainModel

public static GISModel trainModel(ObjectStream<Event> eventStream,
                  int iterations,
                  int cutoff,
                  double sigma)
                           throws IOException

Train a model using the GIS algorithm.

Parameters:

eventStream - The EventStream holding the data on which this model will be trained.

iterations - The number of GIS iterations to perform.

cutoff - The number of times a feature must be seen in order to be relevant for training.

sigma - The standard deviation for the gaussian smoother.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

Throws:

IOException

trainModel

public static GISModel trainModel(int iterations,
                  DataIndexer indexer,
                  boolean smoothing)

Train a model using the GIS algorithm.

Parameters:

iterations - The number of GIS iterations to perform.

indexer - The object which will be used for event compilation.

smoothing - Defines whether the created trainer will use smoothing while training the model.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

trainModel

public static GISModel trainModel(int iterations,
                  DataIndexer indexer)

Train a model using the GIS algorithm.

Parameters:

iterations - The number of GIS iterations to perform.

indexer - The object which will be used for event compilation.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

trainModel

public static GISModel trainModel(int iterations,
                  DataIndexer indexer,
                  Prior modelPrior,
                  int cutoff)

Train a model using the GIS algorithm with the specified number of iterations, data indexer, and prior.

Parameters:

iterations - The number of GIS iterations to perform.

indexer - The object which will be used for event compilation.

modelPrior - The prior distribution for the model.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

trainModel

public static GISModel trainModel(int iterations,
                  DataIndexer indexer,
                  boolean printMessagesWhileTraining,
                  boolean smoothing,
                  Prior modelPrior,
                  int cutoff)

Train a model using the GIS algorithm.

Parameters:

iterations - The number of GIS iterations to perform.

indexer - The object which will be used for event compilation.

printMessagesWhileTraining - Determines whether training status messages are written to STDOUT.

smoothing - Defines whether the created trainer will use smoothing while training the model.

modelPrior - The prior distribution for the model.

cutoff - The number of times a predicate must occur to be used in a model.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.

trainModel

public static GISModel trainModel(int iterations,
                  DataIndexer indexer,
                  boolean printMessagesWhileTraining,
                  boolean smoothing,
                  Prior modelPrior,
                  int cutoff,
                  int threads)

Train a model using the GIS algorithm.

Parameters:

iterations - The number of GIS iterations to perform.

indexer - The object which will be used for event compilation.

printMessagesWhileTraining - Determines whether training status messages are written to STDOUT.

smoothing - Defines whether the created trainer will use smoothing while training the model.

modelPrior - The prior distribution for the model.

cutoff - The number of times a predicate must occur to be used in a model.

Returns:

The newly trained model, which can be used immediately or saved to disk using an opennlp.tools.ml.maxent.io.GISModelWriter object.