classifiersubseteval.java

一个数据挖掘软件ALPHAMINERR的整个过程的JAVA版源代码

/*
 *    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., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

/*
 *    ClassifierSubsetEval.java
 *    Copyright (C) 2000 Mark Hall
 *
 */

package weka.attributeSelection;

import java.io.File;
import java.util.BitSet;
import java.util.Enumeration;
import java.util.Vector;

import weka.classifiers.Classifier;
import weka.classifiers.Evaluation;
import weka.classifiers.rules.ZeroR;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.UnsupportedAttributeTypeException;
import weka.core.Utils;
import weka.filters.Filter;
import weka.filters.unsupervised.attribute.Remove;


/**
 * Classifier subset evaluator. Uses a classifier to estimate the "merit"
 * of a set of attributes.
 *
 * Valid options are:<p>
 *
 * -B <classifier> <br>
 * Class name of the classifier to use for accuracy estimation.
 * Place any classifier options last on the command line following a
 * "--". Eg  -B weka.classifiers.bayes.NaiveBayes ... -- -K <p>
 *
 * -T <br>
 * Use the training data for accuracy estimation rather than a hold out/
 * test set. <p>
 *
 * -H <filename> <br>
 * The file containing hold out/test instances to use for accuracy estimation
 * <p>
 *
 * @author Mark Hall ([e-mail])
 * @version $Revision$
 */
public class ClassifierSubsetEval 
  extends HoldOutSubsetEvaluator
  implements OptionHandler, ErrorBasedMeritEvaluator {

  /** training instances */
  private Instances m_trainingInstances;

  /** class index */
  private int m_classIndex;

  /** number of attributes in the training data */
  private int m_numAttribs;
  
  /** number of training instances */
  private int m_numInstances;

  /** holds the classifier to use for error estimates */
  private Classifier m_Classifier = new ZeroR();

  /** holds the evaluation object to use for evaluating the classifier */
  private Evaluation m_Evaluation;

  /** the file that containts hold out/test instances */
  private File m_holdOutFile = new File("Click to set hold out or "
					+"test instances");

  /** the instances to test on */
  private Instances m_holdOutInstances = null;

  /** evaluate on training data rather than seperate hold out/test set */
  private boolean m_useTraining = false;

  /**
   * Returns a string describing this attribute evaluator
   * @return a description of the evaluator suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return "Evaluates attribute subsets on training data or a seperate "
      +"hold out testing set";
  }

  /**
   * Returns an enumeration describing the available options. <p>
   *
   * -B <classifier> <br>
   * Class name of the classifier to use for accuracy estimation.
   * Place any classifier options last on the command line following a
   * "--". Eg  -B weka.classifiers.bayes.NaiveBayes ... -- -K <p>
   *
   * -T <br>
   * Use the training data for accuracy estimation rather than a hold out/
   * test set. <p>
   *
   * -H <filename> <br>
   * The file containing hold out/test instances to use for accuracy estimation
   * <p>
   *
   * @return an enumeration of all the available options.
   **/
  public Enumeration listOptions () {
    Vector newVector = new Vector(3);
    newVector.addElement(new Option("\tclass name of the classifier to use for" 
				    + "\n\taccuracy estimation. Place any" 
				    + "\n\tclassifier options LAST on the" 
				    + "\n\tcommand line following a \"--\"." 
				    + "\n\teg. -C weka.classifiers.bayes.NaiveBayes ... " 
				    + "-- -K", "B", 1, "-B <classifier>"));
    
    newVector.addElement(new Option("\tUse the training data to estimate"
				    +" accuracy."
				    ,"T",0,"-T"));
    
    newVector.addElement(new Option("\tName of the hold out/test set to "
				    +"\n\testimate accuracy on."
				    ,"H", 1,"-H <filename>"));

    if ((m_Classifier != null) && 
	(m_Classifier instanceof OptionHandler)) {
      newVector.addElement(new Option("", "", 0, "\nOptions specific to " 
				      + "scheme " 
				      + m_Classifier.getClass().getName() 
				      + ":"));
      Enumeration em = ((OptionHandler)m_Classifier).listOptions();

      while (em.hasMoreElements()) {
        newVector.addElement(em.nextElement());
      }
    }

    return  newVector.elements();
  }

  /**
   * Parses a given list of options.
   *
   * Valid options are:<p>
   *
   * -C <classifier> <br>
   * Class name of classifier to use for accuracy estimation.
   * Place any classifier options last on the command line following a
   * "--". Eg  -B weka.classifiers.bayes.NaiveBayes ... -- -K <p>
   *
   * -T <br>
   * Use training data instead of a hold out/test set for accuracy estimation.
   * <p>
   *
   * -H <filname> <br>
   * Name of the hold out/test set to estimate classifier accuracy on.
   * <p>
   *
   * @param options the list of options as an array of strings
   * @exception Exception if an option is not supported
   *
   **/
  public void setOptions (String[] options)
    throws Exception {
    String optionString;
    resetOptions();

    optionString = Utils.getOption('B', options);
    
    if (optionString.length() == 0) {
      throw new Exception("A classifier must be specified with -B option");
    }

    setClassifier(Classifier.forName(optionString,
				     Utils.partitionOptions(options)));

    optionString = Utils.getOption('H',options);
    if (optionString.length() != 0) {
      setHoldOutFile(new File(optionString));
    }

    setUseTraining(Utils.getFlag('T',options));
  }

    /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String classifierTipText() {
    return "Classifier to use for estimating the accuracy of subsets";
  }

  /**
   * Set the classifier to use for accuracy estimation
   *
   * @param newClassifier the Classifier to use.
   */
  public void setClassifier (Classifier newClassifier) {
    m_Classifier = newClassifier;
  }


  /**
   * Get the classifier used as the base learner.
   *
   * @return the classifier used as the classifier
   */
  public Classifier getClassifier () {
    return  m_Classifier;
  }

  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String holdOutFileTipText() {
    return "File containing hold out/test instances.";
  }

  /**
   * Gets the file that holds hold out/test instances.
   * @return File that contains hold out instances
   */
  public File getHoldOutFile() {
    return m_holdOutFile;
  }


  /**
   * Set the file that contains hold out/test instances
   * @param h the hold out file
   */
  public void setHoldOutFile(File h) {
    m_holdOutFile = h;
  }

  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String useTrainingTipText() {
    return "Use training data instead of hold out/test instances.";
  }

  /**
   * Get if training data is to be used instead of hold out/test data
   * @return true if training data is to be used instead of hold out data
   */
  public boolean getUseTraining() {
    return m_useTraining;
  }

  /**
   * Set if training data is to be used instead of hold out/test data
   * @return true if training data is to be used instead of hold out data
   */
  public void setUseTraining(boolean t) {
    m_useTraining = t;
  }

  /**
   * Gets the current settings of ClassifierSubsetEval
   *
   * @return an array of strings suitable for passing to setOptions()
   */
  public String[] getOptions () {
    String[] classifierOptions = new String[0];

    if ((m_Classifier != null) && 
	(m_Classifier instanceof OptionHandler)) {
      classifierOptions = ((OptionHandler)m_Classifier).getOptions();
    }

    String[] options = new String[6 + classifierOptions.length];
    int current = 0;

    if (getClassifier() != null) {
      options[current++] = "-B";
      options[current++] = getClassifier().getClass().getName();
    }

    if (getUseTraining()) {
      options[current++] = "-T";
    }
    options[current++] = "-H"; options[current++] = getHoldOutFile().getPath();
    options[current++] = "--";
    System.arraycopy(classifierOptions, 0, options, current, 
		     classifierOptions.length);
    current += classifierOptions.length;
        while (current < options.length) {
      options[current++] = "";
    }

    return  options;
  }

  /**
   * Generates a attribute evaluator. Has to initialize all fields of the 
   * evaluator that are not being set via options.
   *
   * @param data set of instances serving as training data 
   * @exception Exception if the evaluator has not been 
   * generated successfully
   */
  public void buildEvaluator (Instances data)