net.seninp.jmotif.direct

Class SAXVSMCVErrorFunction

java.lang.Object

net.seninp.jmotif.direct.SAXVSMCVErrorFunction

All Implemented Interfaces:

AbstractErrorFunction

public class SAXVSMCVErrorFunction
extends Object
implements AbstractErrorFunction

This computes the cross-validation classification error given the set of parameters. The direct code was taken from JCOOL (Java COntinuous Optimization Library), and altered for SAX-VSM needs.

See Also:

https://github.com/dhonza/JCOOL/wiki

Field Summary

Fields

Modifier and Type	Field and Description
static Character	DELIMITER

Constructor Summary

Constructors

Constructor and Description
SAXVSMCVErrorFunction(Map<String,List<double[]>> data, int holdOutSampleSize, double nThreshold, net.seninp.jmotif.sax.NumerosityReductionStrategy strategy) Constructor.

Method Summary

Modifier and Type	Method and Description
double	valueAt(Point point) Computes the value at point.

Methods inherited from class java.lang.Object

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

Field Detail

DELIMITER

public static final Character DELIMITER

Constructor Detail

SAXVSMCVErrorFunction

public SAXVSMCVErrorFunction(Map<String,List<double[]>> data,
                             int holdOutSampleSize,
                             double nThreshold,
                             net.seninp.jmotif.sax.NumerosityReductionStrategy strategy)

Constructor.

Parameters:

data - data to use.

holdOutSampleSize - the holdout sample size.

nThreshold - the normalization threshod.

strategy - the nr strategy.

Method Detail

valueAt

public double valueAt(Point point)

Computes the value at point.

Specified by:

valueAt in interface AbstractErrorFunction

Parameters:

point - the point to compute at.

Returns:

computed CV value.