de.unistuttgart.isa.liquidsvm

Class SVM.NPL

java.lang.Object

de.unistuttgart.isa.liquidsvm.SVM

de.unistuttgart.isa.liquidsvm.SVM.NPL

All Implemented Interfaces:

Serializable

Enclosing class:

SVM

public static class SVM.NPL
extends SVM

Neyman-Pearson lemma learning scenario. For the given constraint base several constraints around it are considered given as factors of the original constraint.

Author:

Philipp Thomann

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class de.unistuttgart.isa.liquidsvm.SVM

SVM.EX, SVM.LS, SVM.MC, SVM.NPL, SVM.QT, SVM.ROC

Field Summary

Fields

Modifier and Type	Field and Description
static double[]	DEFAULT_CONSTRAINT_FACTORS

Fields inherited from class de.unistuttgart.isa.liquidsvm.SVM

ENABLE_LOAD_WITHOUT_LIBRARY, MAX_CPU

Constructor Summary

Constructors

Constructor and Description
NPL(double[][] data, double[] labels, double nplClass)
NPL(double[][] data, double[] labels, double nplClass, double constraint)
NPL(double[][] data, double[] labels, double nplClass, double constraint, double[] constraintFactors, Config config)
NPL(LiquidData.Data data, double nplClass)
NPL(LiquidData.Data data, double nplClass, double constraint)
NPL(LiquidData.Data data, double nplClass, double constraint, double[] constraintFactors, Config config)
NPL(LiquidData data, double nplClass)
NPL(LiquidData data, double nplClass, double constraint)
NPL(LiquidData data, double nplClass, double constraint, double[] constraintFactors, Config config)

Method Summary

Modifier and Type	Method and Description
double[][]	select(String... argv) Selects the best hyperparameter pair according to validation error.

Methods inherited from class de.unistuttgart.isa.liquidsvm.SVM

calculateDataCover, clean, finalize, getConfig, getCover, getLastResult, getSelectErrs, getSolutionCoeffs, getSolutionSVs, getTrainErrs, isSelected, isTrained, predict, readSVM, setConfig, setConfig, setConfig, setConfig, setConfigAll, test, test, train

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_CONSTRAINT_FACTORS

public static final double[] DEFAULT_CONSTRAINT_FACTORS

Constructor Detail

NPL

public NPL(LiquidData data,
           double nplClass)

NPL

public NPL(LiquidData data,
           double nplClass,
           double constraint)

NPL

public NPL(LiquidData data,
           double nplClass,
           double constraint,
           double[] constraintFactors,
           Config config)

NPL

public NPL(LiquidData.Data data,
           double nplClass)

NPL

public NPL(LiquidData.Data data,
           double nplClass,
           double constraint)

NPL

public NPL(LiquidData.Data data,
           double nplClass,
           double constraint,
           double[] constraintFactors,
           Config config)

NPL

public NPL(double[][] data,
           double[] labels,
           double nplClass)

NPL

public NPL(double[][] data,
           double[] labels,
           double nplClass,
           double constraint)

NPL

public NPL(double[][] data,
           double[] labels,
           double nplClass,
           double constraint,
           double[] constraintFactors,
           Config config)

Method Detail

select

public double[][] select(String... argv)

Description copied from class: SVM

Selects the best hyperparameter pair according to validation error. argv can be used by experts in the same way as in the svm-select command line program of liquidSVM interface. Most users should rather use Config.

Overrides:

select in class SVM

Parameters:

argv - further command line arguments (for experts)

Returns:

the select errors which can also be retrieved afterwards from SVM.getSelectErrs()