Class TokenNameFinderModel
java.lang.Object
opennlp.tools.util.model.BaseModel
opennlp.tools.namefind.TokenNameFinderModel
- All Implemented Interfaces:
Serializable, ArtifactProvider
The
TokenNameFinderModel is the model used by a learnable TokenNameFinder.- See Also:
-
Nested Class Summary
Nested Classes -
Field Summary
FieldsModifier and TypeFieldDescriptionprivate static final String(package private) static final Stringprivate static final String(package private) static final Stringprivate static final longFields inherited from class BaseModel
artifactMap, FACTORY_NAME, MANIFEST_ENTRY, toolFactory, TRAINING_CUTOFF_PROPERTY, TRAINING_EVENTHASH_PROPERTY, TRAINING_ITERATIONS_PROPERTY -
Constructor Summary
ConstructorsConstructorDescriptionTokenNameFinderModel(File modelFile) Initializes aTokenNameFinderModelinstance via a validFile.Initializes aTokenNameFinderModelinstance via a validInputStream.TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, int beamSize, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, int beamSize, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries, SequenceCodec<String> seqCodec, TokenNameFinderFactory factory) Initializes aTokenNameFinderModelinstance via given parameters.TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.TokenNameFinderModel(String languageCode, SequenceClassificationModel nameFinderModel, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries, SequenceCodec<String> seqCodec, TokenNameFinderFactory factory) Initializes aTokenNameFinderModelinstance via given parameters.TokenNameFinderModel(URL modelURL) Initializes aTokenNameFinderModelinstance via a validURL.TokenNameFinderModel(Path modelPath) Initializes aTokenNameFinderModelinstance via a validPath. -
Method Summary
Modifier and TypeMethodDescriptionstatic Map<String, ArtifactSerializer<?>> Create theserializers.protected voidcreateArtifactSerializers(Map<String, ArtifactSerializer<?>> serializers) Registers allArtifactSerializerfor their artifact file name extensions.protected Class<? extends BaseToolFactory> Subclasses should override this method if their module has a defaultBaseToolFactorysubclass.private voidinit(Object nameFinderModel, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries, SequenceCodec<String> seqCodec) private booleanisModelValid(MaxentModel model) protected voidValidates the parsed artifacts.Methods inherited from class BaseModel
checkArtifactMap, deserialize, getArtifact, getArtifactSerializer, getLanguage, getManifestProperty, getVersion, isLoadedFromSerialized, serialize, serialize, serialize, setManifestProperty, skipEntryForSerialization
-
Field Details
-
serialVersionUID
private static final long serialVersionUID- See Also:
-
COMPONENT_NAME
- See Also:
-
MAXENT_MODEL_ENTRY_NAME
- See Also:
-
GENERATOR_DESCRIPTOR_ENTRY_NAME
- See Also:
-
SEQUENCE_CODEC_CLASS_NAME_PARAMETER
- See Also:
-
-
Constructor Details
-
TokenNameFinderModel
public TokenNameFinderModel(String languageCode, SequenceClassificationModel nameFinderModel, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries, SequenceCodec<String> seqCodec, TokenNameFinderFactory factory) Initializes aTokenNameFinderModelinstance via given parameters.- Parameters:
languageCode- The ISO conform language code.nameFinderModel- A validMaxentModel.generatorDescriptor- Thebyte[]representing the feature generator descriptor.resources- Additional resources in a mapping.manifestInfoEntries- Additional information kept in the manifest.seqCodec- TheSequenceCodecto use.factory- TheTokenNameFinderFactoryfor creating related objects.- Throws:
IllegalArgumentException- Thrown if thenamFinderModelincompatible withseqCodec.
-
TokenNameFinderModel
public TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, int beamSize, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries, SequenceCodec<String> seqCodec, TokenNameFinderFactory factory) Initializes aTokenNameFinderModelinstance via given parameters.- Parameters:
languageCode- The ISO conform language code.nameFinderModel- A validMaxentModel.beamSize- The beam size. Must be greater than0.generatorDescriptor- Thebyte[]representing the feature generator descriptor.resources- Additional resources in a mapping.manifestInfoEntries- Additional information kept in the manifest.seqCodec- TheSequenceCodecto use.factory- TheTokenNameFinderFactoryfor creating related objects.- Throws:
IllegalArgumentException- Thrown if thenamFinderModelincompatible withseqCodec.
-
TokenNameFinderModel
public TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.- Parameters:
languageCode- The ISO conform language code.nameFinderModel- A validMaxentModel.generatorDescriptor- Thebyte[]representing the feature generator descriptor.resources- Additional resources in a mapping.manifestInfoEntries- Additional information kept in the manifest.- Throws:
IllegalArgumentException- Thrown if thenamFinderModelincompatible withseqCodec.
-
TokenNameFinderModel
public TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, int beamSize, byte[] generatorDescriptor, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.- Parameters:
languageCode- The ISO conform language code.nameFinderModel- A validMaxentModel.beamSize- The beam size. Must be greater than0.generatorDescriptor- Thebyte[]representing the feature generator descriptor.resources- Additional resources in a mapping.manifestInfoEntries- Additional information kept in the manifest.- Throws:
IllegalArgumentException- Thrown if thenamFinderModelincompatible withseqCodec.
-
TokenNameFinderModel
public TokenNameFinderModel(String languageCode, MaxentModel nameFinderModel, Map<String, Object> resources, Map<String, String> manifestInfoEntries) Initializes aTokenNameFinderModelinstance via given parameters.- Parameters:
languageCode- The ISO conform language code.nameFinderModel- A validMaxentModel.resources- Additional resources in a mapping.manifestInfoEntries- Additional information kept in the manifest.- Throws:
IllegalArgumentException- Thrown if thenameFinderModelis incompatible withseqCodec.
-
TokenNameFinderModel
Initializes aTokenNameFinderModelinstance via a validInputStream.- Parameters:
in- TheInputStreamused for loading the model.- Throws:
IOException- Thrown if IO errors occurred during initialization.
-
TokenNameFinderModel
Initializes aTokenNameFinderModelinstance via a validFile.- Parameters:
modelFile- TheFileused for loading the model.- Throws:
IOException- Thrown if IO errors occurred during initialization.
-
TokenNameFinderModel
Initializes aTokenNameFinderModelinstance via a validPath.- Parameters:
modelPath- ThePathused for loading the model.- Throws:
IOException- Thrown if IO errors occurred during initialization.
-
TokenNameFinderModel
Initializes aTokenNameFinderModelinstance via a validURL.- Parameters:
modelURL- TheURLused for loading the model.- Throws:
IOException- Thrown if IO errors occurred during initialization.
-
-
Method Details
-
init
-
getNameFinderSequenceModel
- Returns:
- Retrieves a valid
SequenceClassificationModelornullif no matching one could be found.
-
getDefaultFactory
Description copied from class:BaseModelSubclasses should override this method if their module has a defaultBaseToolFactorysubclass.- Overrides:
getDefaultFactoryin classBaseModel- Returns:
- The default
BaseToolFactoryfor the component, ornullif none.
-
getSequenceCodec
- Returns:
- Retrieves the
SequenceCodecin use.
-
getFactory
- Returns:
- Retrieves the
TokenNameFinderFactoryin use.
-
createArtifactSerializers
Description copied from class:BaseModelRegisters allArtifactSerializerfor their artifact file name extensions. The registeredserializersare used to create and serialize resources in the model package.Override this method to register custom
serializers.Note: Subclasses should generally invoke
super.createArtifactSerializersat the beginning of this method.This method is called during construction.
- Overrides:
createArtifactSerializersin classBaseModel- Parameters:
serializers- The key of the map is the file extension used to look up anArtifactSerializer.
-
createArtifactSerializers
Create theserializers. Currently, for serializers related to features that require external resources, such asW2VClassesDictionaryobjects, the convention is to add its element tag name as key of the serializer map.For example, the element tag name for the
WordClusterFeatureGeneratorwhich usesW2VClassesDictionaryobjects serialized by theW2VClassesDictionarySerializeris 'wordcluster', which is the key used to add the serializer to the map.- Returns:
- A
Mapcontaining the addedserializers.
-
isModelValid
-
validateArtifactMap
Description copied from class:BaseModelValidates the parsed artifacts. If something is not valid, subclasses should throw anInvalidFormatException.Note: Subclasses should generally invoke
super.validateArtifactMapat the beginning of this method.- Overrides:
validateArtifactMapin classBaseModel- Throws:
InvalidFormatException- Thrown if artifacts were found to be inconsistent.
-