|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Object opennlp.tools.namefind.TokenNameFinderCrossValidator
public class TokenNameFinderCrossValidator
Method Summary | |
---|---|
void |
evaluate(ObjectStream<NameSample> samples,
int nFolds)
Starts the evaluation. |
FMeasure |
getFMeasure()
|
Methods inherited from class java.lang.Object |
---|
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Constructor Detail |
---|
@Deprecated public TokenNameFinderCrossValidator(String languageCode, int cutoff, int iterations)
TokenNameFinderCrossValidator(String, String, TrainingParameters, byte[], Map, TokenNameFinderEvaluationMonitor...)
instead and pass in a TrainingParameters object.
languageCode
- the language of the training datacutoff
- iterations
- @Deprecated public TokenNameFinderCrossValidator(String languageCode, String type, int cutoff, int iterations)
TokenNameFinderCrossValidator(String, String, TrainingParameters, byte[], Map, TokenNameFinderEvaluationMonitor...)
instead and pass in a TrainingParameters object.
languageCode
- the language of the training datatype
- null or an override type for all types in the training datacutoff
- specifies the min number of times a feature must be seeniterations
- the number of iterations@Deprecated public TokenNameFinderCrossValidator(String languageCode, String type, byte[] featureGeneratorBytes, Map<String,Object> resources, int iterations, int cutoff)
TokenNameFinderCrossValidator(String, String, TrainingParameters, byte[], Map, TokenNameFinderEvaluationMonitor...)
instead and pass in a TrainingParameters object.
languageCode
- the language of the training datatype
- null or an override type for all types in the training datafeatureGeneratorBytes
- descriptor to configure the feature generation or nullresources
- the resources for the name finder or null if nonecutoff
- specifies the min number of times a feature must be seeniterations
- the number of iterationspublic TokenNameFinderCrossValidator(String languageCode, String type, TrainingParameters trainParams, byte[] featureGeneratorBytes, Map<String,Object> resources, TokenNameFinderEvaluationMonitor... listeners)
languageCode
- the language of the training datatype
- null or an override type for all types in the training datatrainParams
- machine learning train parametersfeatureGeneratorBytes
- descriptor to configure the feature generation or nulllisteners
- a list of listenersresources
- the resources for the name finder or null if noneMethod Detail |
---|
public void evaluate(ObjectStream<NameSample> samples, int nFolds) throws IOException
samples
- the data to train and testnFolds
- number of folds
IOException
public FMeasure getFMeasure()
|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |