mulan.classifier
Interface MultiLabelLearner

All Known Implementing Classes:

AdaBoostMH, BinaryRelevance, BPMLL, BRkNN, CalibratedLabelRanking, ClassifierChain, ClusteringBased, EnsembleOfClassifierChains, EnsembleOfPrunedSets, EnsembleOfSubsetLearners, ExampleBasedFMeasureOptimizer, HMC, HMCNode, HOMER, IBLR_ML, IncludeLabelsClassifier, LabelPowerset, LabelsetPruning, Meta, MetaLabeler, MLkNN, MLPTO, MMPLearner, MultiClassLearner, MultiLabelKNN, MultiLabelLearnerBase, MultiLabelMetaLearner, MultiLabelStacking, OneThreshold, PPT, PrunedSets, RAkEL, RAkELd, RCut, SCut, SubsetLearner, ThresholdPrediction, TransformationBasedMultiLabelLearner

public interface MultiLabelLearner

Common root interface for all multi-label learner types.

Author:

Jozef Vilcek

Method Summary

void	build(MultiLabelInstances instances) Builds the learner model from specified MultiLabelInstances data.
boolean	isUpdatable() Returns value indicating if learner is updatable, so if learner is able to perform on-line learning.
MultiLabelLearner	makeCopy() Creates a deep copy of the given learner using serialization.
MultiLabelOutput	makePrediction(Instance instance) Returns the prediction of the learner for a given input Instance.
void	setDebug(boolean debug) Sets whether debugging information should be output by the model

Method Detail

isUpdatable

boolean isUpdatable()

Returns value indicating if learner is updatable, so if learner is able to perform on-line learning. The fact if learner is updatable or not influences the behavior of build(MultiLabelInstances) method.

When false is returned, each call of the build(MultiLabelInstances) will initialize the learner from the scratch, removing any potential knowledge built by previously entered training data. When true is returned, then on the first call of the build(MultiLabelInstances) the learner is initialized with the passed training data. All other calls contribute to the existing learner's model with new data.

Returns:

true if learner is updatable (on-line), false otherwise.

build

void build(MultiLabelInstances instances)
           throws Exception,
                  InvalidDataException

Builds the learner model from specified MultiLabelInstances data. Sequential calls to this method either re-build the learners model with new data (off-line learner) or contribute to the existing model with new data (on-line learner). The behavior is determined by the outcome of isUpdatable() method.

Parameters:

instances - set of training data, upon which the learner model should be built

Throws:

Exception - if learner model was not created successfully

InvalidDataException - if specified instances data is invalid or not supported by the learner

See Also:

isUpdatable()

makeCopy

MultiLabelLearner makeCopy()
                           throws Exception

Creates a deep copy of the given learner using serialization.

Returns:

a deep copy of the learner

Throws:

Exception - if an error occurs while making copy of the learner.

makePrediction

MultiLabelOutput makePrediction(Instance instance)
                                throws Exception,
                                       InvalidDataException,
                                       ModelInitializationException

Returns the prediction of the learner for a given input Instance.

Parameters:

instance - the input given to the learner in the form of Instance

Returns:

a prediction of the learner in form of MultiLabelOutput.

Throws:

Exception - if an error occurs while making the prediction.

InvalidDataException - if specified instance data is invalid and can not be processed by the learner

ModelInitializationException - if method is called before build(MultiLabelInstances)

setDebug

void setDebug(boolean debug)

Sets whether debugging information should be output by the model

Parameters:

debug -