normalization.java

一个很好的LIBSVM的JAVA源码。对于要研究和改进SVM算法的学者。可以参考。来自数据挖掘工具YALE工具包。

/*
 *  YALE - Yet Another Learning Environment
 *  Copyright (C) 2001-2004
 *      Simon Fischer, Ralf Klinkenberg, Ingo Mierswa, 
 *          Katharina Morik, Oliver Ritthoff
 *      Artificial Intelligence Unit
 *      Computer Science Department
 *      University of Dortmund
 *      44221 Dortmund,  Germany
 *  email: [e-mail]
 *  web:   http://yale.cs.uni-dortmund.de/
 *
 *  This program 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 2 of the
 *  License, or (at your option) any later version. 
 *
 *  This program 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, write to the Free Software
 *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
 *  USA.
 */
package edu.udo.cs.yale.operator.preprocessing;

import edu.udo.cs.yale.operator.Operator;
import edu.udo.cs.yale.operator.IOObject;
import edu.udo.cs.yale.operator.OperatorException;
import edu.udo.cs.yale.operator.UserError;
import edu.udo.cs.yale.operator.parameter.*;
import edu.udo.cs.yale.generator.GenerationException;
import edu.udo.cs.yale.tools.LogService;
import edu.udo.cs.yale.tools.ParameterService;
import edu.udo.cs.yale.tools.Ontology;
import edu.udo.cs.yale.example.Attribute;
import edu.udo.cs.yale.example.AttributeParser;
import edu.udo.cs.yale.example.ExampleSet;
import edu.udo.cs.yale.example.ExampleReader;
import edu.udo.cs.yale.example.Example;
import edu.udo.cs.yale.generator.*;

import java.io.File;
import java.io.FileReader;
import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;
import java.util.ListIterator;

/** This operator performs a normalization.
 *
 *  @version $Id: Normalization.java,v 1.5 2004/09/09 12:00:53 ingomierswa Exp $
 */
public class Normalization extends Operator {

    private static final Class[] INPUT_CLASSES  = { ExampleSet.class };
    private static final Class[] OUTPUT_CLASSES = { ExampleSet.class };

    public Class[] getInputClasses() { return INPUT_CLASSES; }
    public Class[] getOutputClasses() { return OUTPUT_CLASSES; }


    public IOObject[] apply() throws OperatorException {
	ExampleSet exampleSet = (ExampleSet)getInput(ExampleSet.class);
        
        if (!getParameterAsBoolean("mean_variance_scaling")) {
            LogService.logMessage("Mean-variance-scaling set to OFF",LogService.MINIMUM);
            ArrayList generators = new ArrayList();
	    
            double min = getParameterAsDouble("min");
            double max = getParameterAsDouble("max");
            if (max < min) 
		throw new UserError(this, 116, "max", "Must not be smaller than 'min'");
	    for (int i = 0; i < exampleSet.getNumberOfAttributes(); i++) {
		FeatureGenerator g =  new NormalizationGenerator(min, max);
                g.setArguments(new Attribute[] { exampleSet.getAttribute(i) });
                generators.add(g);
	    }
	    
            try {
                List attributes = FeatureGenerator.generateAll(exampleSet.getExampleTable(),
							       generators);
                exampleSet.removeAllAttributes();
                exampleSet.addAllAttributes(attributes);
            } catch (GenerationException e) {
		throw new UserError(this, e, 108, e.getMessage());
	    }
        } else {
            LogService.logMessage("Mean-variance-scaling set to ON",LogService.MINIMUM);
            ExampleReader r = exampleSet.getExampleReader();
            while (r.hasNext()) {
		Example example = (Example)r.next();
		for (int i = 0; i < exampleSet.getNumberOfAttributes(); i++) {
		    Attribute attribute = exampleSet.getAttribute(i);
		    if (attribute.getVariance() == 0) {
			example.setValue(attribute, 0);
		    } else {
			double newValue = (example.getValue(attribute)-attribute.getAverage())/(Math.sqrt(attribute.getVariance()));
			example.setValue(attribute, newValue);
		    }
		}
            }
        }
        exampleSet.recalculateAllAttributeStatistics();
        return new IOObject[] { exampleSet };
    }
    
    public List getParameterTypes() {
	List types = super.getParameterTypes();
	types.add(new ParameterTypeDouble("min", "The minimum value after normalization", 
					  Double.NEGATIVE_INFINITY, 
					  Double.POSITIVE_INFINITY, 0.0d));
	types.add(new ParameterTypeDouble("max", "The maximum value after normalization", 
					  Double.NEGATIVE_INFINITY, 
					  Double.POSITIVE_INFINITY, 1.0d));
	ParameterType type = 
	    new ParameterTypeBoolean("mean_variance_scaling", 
				     "Determines whether to perform mean-variance-scaling or not; scaling ignores min- and max-setings", 
				     true);
	type.setExpert(false);
        types.add(type);
	return types;
    }

}