Package se.hb.jcp.cp

Class InductiveConformalClassifier

java.lang.Object

se.hb.jcp.cp.InductiveConformalClassifier

All Implemented Interfaces:

java.io.Serializable, IConformalClassifier, IClassifierInformation

public class InductiveConformalClassifier
extends java.lang.Object
implements IConformalClassifier, java.io.Serializable

Represents an instance of a specific inductive conformal classification algorithm.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor	Description
InductiveConformalClassifier(IClassificationNonconformityFunction nc, double[] targets)	Creates an inductive conformal classifier using the supplied information.
InductiveConformalClassifier(IClassificationNonconformityFunction nc, double[] targets, boolean useLabelConditionalCP)	Creates an inductive conformal classifier using the supplied information.

Method Summary

Modifier and Type	Method	Description
void	calibrate(cern.colt.matrix.DoubleMatrix2D xcal, double[] ycal)	Calibrates this conformal classifier using the supplied data.
void	fit(cern.colt.matrix.DoubleMatrix2D xtr, double[] ytr, cern.colt.matrix.DoubleMatrix2D xcal, double[] ycal)	Trains and calibrates this conformal classifier using the supplied data.
int	getAttributeCount()	Returns the number of attributes the classifier has been trained on.
java.lang.Double[]	getLabels()	Returns the set of class labels the classifier has been trained on.
IClassificationNonconformityFunction	getNonconformityFunction()	Returns the associated nonconformity function.
boolean	isTrained()	Returns whether this classifier has been trained and calibrated.
cern.colt.matrix.DoubleMatrix1D	nativeStorageTemplate()	Returns a value of the DoubleMatrix1D derived class that is the native storage format for the classifier.
ConformalClassification	predict(cern.colt.matrix.DoubleMatrix1D x)	Makes a prediction for the instance x.
ConformalClassification[]	predict(cern.colt.matrix.DoubleMatrix2D x)	Makes a prediction for each instance in x.
cern.colt.matrix.DoubleMatrix1D	predictPValues(cern.colt.matrix.DoubleMatrix1D x)	Computes the predicted p-values for the instance x.
void	predictPValues(cern.colt.matrix.DoubleMatrix1D x, cern.colt.matrix.DoubleMatrix1D pValues)	Computes the predicted p-values for the instance x.
cern.colt.matrix.DoubleMatrix2D	predictPValues(cern.colt.matrix.DoubleMatrix2D x)	Computes the predicted p-values for each target and instance in x.
void	setNonconformityFunction(IClassificationNonconformityFunction nc)	Sets a new non-conformity function in this inductive conformal classifier.

Methods inherited from class java.lang.Object

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

Constructor Detail

InductiveConformalClassifier

public InductiveConformalClassifier(IClassificationNonconformityFunction nc,
                                    double[] targets)

Creates an inductive conformal classifier using the supplied information.

Parameters:

nc - the untrained non-conformity function to use.

targets - the class labels.

InductiveConformalClassifier

public InductiveConformalClassifier(IClassificationNonconformityFunction nc,
                                    double[] targets,
                                    boolean useLabelConditionalCP)

Creates an inductive conformal classifier using the supplied information.

Parameters:

nc - the untrained non-conformity function to use.

targets - the class labels.

useLabelConditionalCP - a boolean indicating whether label conditional conformal prediction should be used.

Method Detail

fit

public void fit(cern.colt.matrix.DoubleMatrix2D xtr,
                double[] ytr,
                cern.colt.matrix.DoubleMatrix2D xcal,
                double[] ycal)

Trains and calibrates this conformal classifier using the supplied data.

Parameters:

xtr - the attributes of the training instances.

ytr - the targets of the training instances.

xcal - the attributes of the calibration instances.

ycal - the targets of the calibration instances.

calibrate

public void calibrate(cern.colt.matrix.DoubleMatrix2D xcal,
                      double[] ycal)

Calibrates this conformal classifier using the supplied data. The classifier's non-conformity function must have been trained first.

Parameters:

xcal - the attributes of the calibration instances.

ycal - the targets of the calibration instances.

predict

public ConformalClassification[] predict(cern.colt.matrix.DoubleMatrix2D x)

Makes a prediction for each instance in x. The method is parallellized over the instances.

Specified by:

predict in interface IConformalClassifier

Parameters:

x - the instances.

Returns:

an array containing a ConformalClassification for each instance.

predict

public ConformalClassification predict(cern.colt.matrix.DoubleMatrix1D x)

Makes a prediction for the instance x.

Specified by:

predict in interface IConformalClassifier

Parameters:

x - the instance.

Returns:

a prediction in the form of a ConformalClassification.

predictPValues

public cern.colt.matrix.DoubleMatrix2D predictPValues(cern.colt.matrix.DoubleMatrix2D x)

Computes the predicted p-values for each target and instance in x. The method is parallellized over the instances.

Specified by:

predictPValues in interface IConformalClassifier

Parameters:

x - the instances.

Returns:

an DoubleMatrix2D containing the predicted p-values for each instance.

predictPValues

public cern.colt.matrix.DoubleMatrix1D predictPValues(cern.colt.matrix.DoubleMatrix1D x)

Computes the predicted p-values for the instance x.

Specified by:

predictPValues in interface IConformalClassifier

Parameters:

x - the instance.

Returns:

an DoubleMatrix1D containing the predicted p-values.

predictPValues

public void predictPValues(cern.colt.matrix.DoubleMatrix1D x,
                           cern.colt.matrix.DoubleMatrix1D pValues)

Computes the predicted p-values for the instance x.

Specified by:

predictPValues in interface IConformalClassifier

Parameters:

x - the instance.

pValues - an initialized DoubleMatrix1D to store the p-values.

getNonconformityFunction

public IClassificationNonconformityFunction getNonconformityFunction()

Description copied from interface: IConformalClassifier

Returns the associated nonconformity function.

Specified by:

getNonconformityFunction in interface IConformalClassifier

Returns:

the associated IClassificationNonconformityFunction.

setNonconformityFunction

public void setNonconformityFunction(IClassificationNonconformityFunction nc)

Sets a new non-conformity function in this inductive conformal classifier. The conformal classifier must be (re)calibrated before the new non-conformity function can be used for predictions.

Parameters:

nc - the new non-conformity function.

isTrained

public boolean isTrained()

Returns whether this classifier has been trained and calibrated.

Specified by:

isTrained in interface IClassifierInformation

Returns:

true if the classifier has been trained and calibrated or false otherwise.

getAttributeCount

public int getAttributeCount()

Description copied from interface: IClassifierInformation

Returns the number of attributes the classifier has been trained on.

Specified by:

getAttributeCount in interface IClassifierInformation

Returns:

Returns the number of attributes the classifier has been trained on or -1 if the classifier has not been trained.

getLabels

public java.lang.Double[] getLabels()

Description copied from interface: IClassifierInformation

Returns the set of class labels the classifier has been trained on.

Specified by:

getLabels in interface IClassifierInformation

Returns:

Returns the class labels the classifier has been trained on in increasing order (the index is the class id) or null if the classifier has not been trained.

nativeStorageTemplate

public cern.colt.matrix.DoubleMatrix1D nativeStorageTemplate()

Description copied from interface: IClassifierInformation

Returns a value of the DoubleMatrix1D derived class that is the native storage format for the classifier.

Specified by:

nativeStorageTemplate in interface IClassifierInformation

Returns:

a value of the DoubleMatrix1D derived class of the native storage format for the classifier.