opennlp.tools.doccat

Class DocumentCategorizerME

java.lang.Object

opennlp.tools.doccat.DocumentCategorizerME

All Implemented Interfaces:

DocumentCategorizer

public class DocumentCategorizerME
extends Object
implements DocumentCategorizer

Maxent implementation of DocumentCategorizer.

Constructor Summary

Constructors

Constructor and Description
DocumentCategorizerME(DoccatModel model) Initializes the current instance with a doccat model.
DocumentCategorizerME(DoccatModel model, FeatureGenerator... featureGenerators) Initializes a the current instance with a doccat model and custom feature generation.
DocumentCategorizerME(MaxentModel model) Deprecated. Use DocumentCategorizerME(DoccatModel) instead.
DocumentCategorizerME(MaxentModel model, FeatureGenerator... featureGenerators) Deprecated. Use DocumentCategorizerME(DoccatModel, FeatureGenerator...) instead.

Method Summary

Modifier and Type	Method and Description
double[]	categorize(String documentText) Categorizes the given text.
double[]	categorize(String[] text) Categorizes the given text.
String	getAllResults(double[] results)
String	getBestCategory(double[] outcome)
String	getCategory(int index)
int	getIndex(String category)
int	getNumberOfCategories()
static AbstractModel	train(DocumentCategorizerEventStream eventStream) Deprecated.
static DoccatModel	train(String languageCode, ObjectStream<DocumentSample> samples) Trains a doccat model with default feature generation.
static DoccatModel	train(String languageCode, ObjectStream<DocumentSample> samples, int cutoff, int iterations) Trains a doccat model with default feature generation.
static DoccatModel	train(String languageCode, ObjectStream<DocumentSample> samples, int cutoff, int iterations, FeatureGenerator... featureGenerators) Trains a document categorizer model with custom feature generation.
static DoccatModel	train(String languageCode, ObjectStream<DocumentSample> samples, TrainingParameters mlParams, FeatureGenerator... featureGenerators)

Methods inherited from class java.lang.Object

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

Constructor Detail

DocumentCategorizerME

public DocumentCategorizerME(DoccatModel model,
                             FeatureGenerator... featureGenerators)

Initializes a the current instance with a doccat model and custom feature generation. The feature generation must be identical to the configuration at training time.

Parameters:

model -

featureGenerators -

DocumentCategorizerME

public DocumentCategorizerME(DoccatModel model)

Initializes the current instance with a doccat model. Default feature generation is used.

Parameters:

model -

DocumentCategorizerME

@Deprecated
public DocumentCategorizerME(MaxentModel model)

Deprecated. Use DocumentCategorizerME(DoccatModel) instead.

Initializes the current instance with the given MaxentModel.

Parameters:

model -

DocumentCategorizerME

@Deprecated
public DocumentCategorizerME(MaxentModel model,
                                         FeatureGenerator... featureGenerators)

Deprecated. Use DocumentCategorizerME(DoccatModel, FeatureGenerator...) instead.

Initializes the current instance with a the given MaxentModel and FeatureGenerators.

Parameters:

model -

featureGenerators -

Method Detail

categorize

public double[] categorize(String[] text)

Categorizes the given text.

Specified by:

categorize in interface DocumentCategorizer

Parameters:

text -

categorize

public double[] categorize(String documentText)

Categorizes the given text. The text is tokenized with the SimpleTokenizer before it is passed to the feature generation.

Specified by:

categorize in interface DocumentCategorizer

getBestCategory

public String getBestCategory(double[] outcome)

Specified by:

getBestCategory in interface DocumentCategorizer

getIndex

public int getIndex(String category)

Specified by:

getIndex in interface DocumentCategorizer

getCategory

public String getCategory(int index)

Specified by:

getCategory in interface DocumentCategorizer

getNumberOfCategories

public int getNumberOfCategories()

Specified by:

getNumberOfCategories in interface DocumentCategorizer

getAllResults

public String getAllResults(double[] results)

Specified by:

getAllResults in interface DocumentCategorizer

train

@Deprecated
public static AbstractModel train(DocumentCategorizerEventStream eventStream)
                                       throws IOException

Deprecated.

Trains a new model for the DocumentCategorizerME.

Parameters:

eventStream -

Returns:

the new model

Throws:

IOException

train

public static DoccatModel train(String languageCode,
                                ObjectStream<DocumentSample> samples,
                                TrainingParameters mlParams,
                                FeatureGenerator... featureGenerators)
                         throws IOException

Throws:

IOException

train

public static DoccatModel train(String languageCode,
                                ObjectStream<DocumentSample> samples,
                                int cutoff,
                                int iterations,
                                FeatureGenerator... featureGenerators)
                         throws IOException

Trains a document categorizer model with custom feature generation.

Parameters:

languageCode -

samples -

cutoff -

iterations -

featureGenerators -

Returns:

the trained doccat model

Throws:

IOException

train

public static DoccatModel train(String languageCode,
                                ObjectStream<DocumentSample> samples,
                                int cutoff,
                                int iterations)
                         throws IOException

Trains a doccat model with default feature generation.

Parameters:

languageCode -

samples -

Returns:

the trained doccat model

Throws:

IOException

ObjectStreamException

train

public static DoccatModel train(String languageCode,
                                ObjectStream<DocumentSample> samples)
                         throws IOException

Trains a doccat model with default feature generation.

Parameters:

languageCode -

samples -

Returns:

the trained doccat model

Throws:

IOException

ObjectStreamException