uk.ac.ed.inf.enzml.mulan.predict
Class MulanPredict

java.lang.Object
  uk.ac.ed.inf.utils.Initialised
      uk.ac.ed.inf.enzml.mulan.predict.MulanPredict

All Implemented Interfaces:

uk.ac.ed.inf.utils.Initialisable

Direct Known Subclasses:

MulanPredictWithTrainedModel

public class MulanPredict
extends uk.ac.ed.inf.utils.Initialised

First trains on a dataset (training set) and then emits predictions for a (compatible unlabelled) dataset.

Version:

19 Sep 2010

Author:

Luna De Ferrari luna.deferrari-at-ed.ac.uk

Field Summary

static java.lang.String	ATTRIBUTE_ID
static java.lang.String	ATTRIBUTE_NAME
static java.lang.String	CONFIDENCE
static java.lang.String	INSTANCE_ID
static java.lang.String	INSTANCE_NAME

Fields inherited from class uk.ac.ed.inf.utils.Initialised

m_initialisationIsCorrect

Constructor Summary

MulanPredict(int trainArffId, int unlabeledArffId, java.lang.String arffRecordsDatabase, java.lang.String predictionsDirectory, MulanLearner learner)

Method Summary

java.util.Vector<java.lang.String>	generateInstancesNames()
java.util.TreeSet<java.lang.String>	getInstances(uk.ac.ed.inf.utils.database.DbReader dataReader, java.lang.String instancesQuery)
java.lang.String	getPredictionsFile() The full path to the predictions file
MulanArffRecord	getTrainArff()
MulanArffRecord	getUnlabeledArff()
weka.core.Instances	getUnlabeledData()
static void	main(java.lang.String[] args)
void	predict()
static java.lang.String	predictionsHeader()
java.lang.String	processOutput(int instanceIndex, mulan.classifier.MultiLabelOutput instanceOutput)
static java.lang.String	run(int trainId, int testId, java.lang.String machineLearningProps, java.lang.String predictionsDirectory, MulanLearner learner) Generates and saves predictions.
void	savePredictions()
void	setPredictionsAndLogFile(java.lang.String predictionsDirectory)

Methods inherited from class uk.ac.ed.inf.utils.Initialised

initialisationIsCorrect, setInitialisationIsCorrect

Methods inherited from class java.lang.Object

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

Field Detail

INSTANCE_ID

public static final java.lang.String INSTANCE_ID

See Also:

Constant Field Values

INSTANCE_NAME

public static final java.lang.String INSTANCE_NAME

See Also:

Constant Field Values

ATTRIBUTE_ID

public static final java.lang.String ATTRIBUTE_ID

See Also:

Constant Field Values

ATTRIBUTE_NAME

public static final java.lang.String ATTRIBUTE_NAME

See Also:

Constant Field Values

CONFIDENCE

public static final java.lang.String CONFIDENCE

See Also:

Constant Field Values

Constructor Detail

MulanPredict

public MulanPredict(int trainArffId,
                    int unlabeledArffId,
                    java.lang.String arffRecordsDatabase,
                    java.lang.String predictionsDirectory,
                    MulanLearner learner)

Method Detail

generateInstancesNames

public java.util.Vector<java.lang.String> generateInstancesNames()

getInstances

public java.util.TreeSet<java.lang.String> getInstances(uk.ac.ed.inf.utils.database.DbReader dataReader,
                                                        java.lang.String instancesQuery)

getPredictionsFile

public java.lang.String getPredictionsFile()

The full path to the predictions file

Returns:

full path to the predictions file

getTrainArff

public MulanArffRecord getTrainArff()

getUnlabeledArff

public MulanArffRecord getUnlabeledArff()

getUnlabeledData

public weka.core.Instances getUnlabeledData()

predict

public void predict()

processOutput

public java.lang.String processOutput(int instanceIndex,
                                      mulan.classifier.MultiLabelOutput instanceOutput)

savePredictions

public void savePredictions()

setPredictionsAndLogFile

public void setPredictionsAndLogFile(java.lang.String predictionsDirectory)

main

public static void main(java.lang.String[] args)

predictionsHeader

public static java.lang.String predictionsHeader()

run

public static java.lang.String run(int trainId,
                                   int testId,
                                   java.lang.String machineLearningProps,
                                   java.lang.String predictionsDirectory,
                                   MulanLearner learner)

Generates and saves predictions.

Parameters:

trainId -

testId -

machineLearningProps -

predictionsDirectory -

learner -

Returns:

the path to the predictions file