randomsplitvalidationchain.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.validation;

import edu.udo.cs.yale.operator.IOObject;
import edu.udo.cs.yale.operator.IOContainer;
import edu.udo.cs.yale.operator.OperatorException;
import edu.udo.cs.yale.operator.parameter.*;
import edu.udo.cs.yale.tools.LogService;
import edu.udo.cs.yale.tools.math.AverageVector;
import edu.udo.cs.yale.example.ExampleSet;
import edu.udo.cs.yale.example.SplittedExampleSet;
import edu.udo.cs.yale.operator.learner.Model;
import edu.udo.cs.yale.operator.performance.PerformanceVector;
import edu.udo.cs.yale.operator.performance.PerformanceCriterion;

import java.util.List;
import java.util.LinkedList;

/** A <code>RandomSplitValidationChain</code> splits up the example set into a training and test
 *  set and evaluates the model. 
 *  The first inner operator must accept an
 *  {@link edu.udo.cs.yale.example.ExampleSet} while the second must accept an
 *  {@link edu.udo.cs.yale.example.ExampleSet} and the output of the first (which is
 *  in most cases a {@link edu.udo.cs.yale.operator.learner.Model}) and must produce
 *  a {@link edu.udo.cs.yale.operator.performance.PerformanceVector}.
 *
 *  @yale.xmlclass RandomSplitValidationChain *
 *  @author simon, ingo
 *  @version $Id: RandomSplitValidationChain.java,v 1.4 2004/10/07 20:31:00 ingomierswa Exp $
 */
public class RandomSplitValidationChain extends ValidationChain { 

    public IOObject[] apply() throws OperatorException {
	double splitRatio = getParameterAsDouble("split_ratio");
	SplittedExampleSet eSet = new SplittedExampleSet((ExampleSet)getInput(ExampleSet.class), splitRatio);

	eSet.selectSingleSubset(0);
	learn(eSet);
	eSet.selectSingleSubset(1);
	IOContainer evalRes = evaluate(eSet);
	List averageVectors = new LinkedList();
	handleAverages(evalRes, averageVectors);
	PerformanceVector performanceVector = getPerformanceVector(averageVectors);
	if (performanceVector != null)
	    setResult(performanceVector.getMainCriterion());

	AverageVector[] result = new AverageVector[averageVectors.size()];
	averageVectors.toArray(result);
	return result;
    }
    
    public int getNumberOfValidationSteps() {
	return 1;
    }

    public List getParameterTypes() {
	List types = super.getParameterTypes();
	ParameterType type = new ParameterTypeDouble("split_ratio", "Relative size of the training set.", 0.0d, 1.0d, 0.7d);
	type.setExpert(false);
	types.add(type);
	return types;
    }
}