/*
* $RCSfile: MetricFactory.java,v $
* $Revision: 1.6 $
* $Date: 2008/01/08 14:33:15 $
* $Author: wojna $
*
* Copyright (C) 2002 - 2007 Logic Group, Institute of Mathematics, Warsaw University
*
* This file is part of Rseslib.
*
* Rseslib is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* Rseslib is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
package rseslib.processing.metrics;
import java.util.Properties;
import rseslib.structure.metric.CityHammingMetric;
import rseslib.structure.metric.CitySVDMetric;
import rseslib.structure.metric.InterpolatedVDMetric;
import rseslib.structure.metric.DensityBasedVDMetric;
import rseslib.structure.metric.Metric;
import rseslib.structure.metric.AbstractWeightedMetric;
import rseslib.structure.table.DoubleDataTable;
import rseslib.system.Configuration;
import rseslib.system.PropertyConfigurationException;
import rseslib.system.progress.Progress;
/**
* Factory of metrics inducing a required metric
* from a training table.
*
* @author Arkadiusz Wojna
*/
public class MetricFactory
{
/** Property name for metric type. */
public static final String METRIC_PROPERTY_NAME = "metric";
/** Property name for metric type. */
public static final String VICINITY_SIZE_FOR_DBVDM_PROPERTY_NAME = "vicinitySizeForDensityBasedMetric";
/** Metric types. */
public enum MetricType
{
CityAndHamming(CityHammingMetric.class),
CityAndSimpleValueDifference(CitySVDMetric.class),
DensityBasedValueDifference(DensityBasedVDMetric.class),
InterpolatedValueDifference(InterpolatedVDMetric.class);
/** The class of this metric type. */
private final Class classobj;
/** Constructor. */
MetricType(Class classobj)
{
this.classobj = classobj;
}
}
/** Attribute weighting methods. */
public enum Weighting { None, Perceptron, DistanceBased, AccuracyBased; }
/**
* Induces the appropriate metric from a given training set of data object.
* @param tab The set of data objects used to induce this metric.
* @param metricName The name of a metric type.
*
* @return The induced metric.
*/
public static Metric getMetric(Properties metricProperties, DoubleDataTable tab) throws PropertyConfigurationException
{
MetricType metric;
try
{
metric = MetricType.valueOf(metricProperties.getProperty(METRIC_PROPERTY_NAME));
}
catch (IllegalArgumentException e)
{
throw new PropertyConfigurationException("Unknown metric: "+metricProperties.getProperty(METRIC_PROPERTY_NAME));
}
Properties metricProp = Configuration.loadDefaultProperties(metric.classobj);
switch (metric)
{
case CityAndHamming:
return new CityHammingMetric(metricProp, tab);
case CityAndSimpleValueDifference:
return new CitySVDMetric(metricProp, tab);
case DensityBasedValueDifference:
metricProp.setProperty(DensityBasedVDMetric.VICINITY_SIZE_PROPERTY_NAME, metricProperties.getProperty(VICINITY_SIZE_FOR_DBVDM_PROPERTY_NAME));
return new DensityBasedVDMetric(metricProp, tab);
case InterpolatedValueDifference:
return new InterpolatedVDMetric(metricProp, tab);
}
return null;
}
/**
* Calls the appropriate weighting method.
*
* @param prop Properties with the weighting method.
* @param metr Original metric to be weighted.
* @param tab Training set of data objects.
* @param prog Progress object.
* @throws InterruptedException when the user interrupts the execution.
*/
public static void adjustWeights(String weightingMethodName, AbstractWeightedMetric metr, DoubleDataTable tab, Progress prog) throws PropertyConfigurationException, InterruptedException
{
Weighting weightingMethod;
try
{
weightingMethod = Weighting.valueOf(weightingMethodName);
}
catch (IllegalArgumentException e)
{
throw new PropertyConfigurationException("Unknown weighting method: "+weightingMethodName);
}
WeightAdjuster weightAdj = null;
switch (weightingMethod)
{
case None:
prog.set("Weighting attributes in a metric", 1);
prog.step();
break;
case Perceptron:
weightAdj = new PerceptronBasedWeightAdjuster(null);
break;
case DistanceBased:
weightAdj = new DistanceBasedWeightAdjuster(null);
break;
case AccuracyBased:
weightAdj = new AccuracyBasedWeightAdjuster(null);
break;
}
if (weightAdj!=null)
{
weightAdj.adjustWeights(metr, tab, prog);
}
return;
}
}