midd.java

代码是一个分类器的实现,其中使用了部分weka的源代码。可以将项目导入eclipse运行

/*
 *    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.
 */

/*
 * MIDD.java
 * Copyright (C) 2005 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.classifiers.mi;

import weka.classifiers.Classifier;
import weka.core.Capabilities;
import weka.core.FastVector;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.MultiInstanceCapabilitiesHandler;
import weka.core.Optimization;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.SelectedTag;
import weka.core.Tag;
import weka.core.TechnicalInformation;
import weka.core.TechnicalInformationHandler;
import weka.core.Utils;
import weka.core.Capabilities.Capability;
import weka.core.TechnicalInformation.Field;
import weka.core.TechnicalInformation.Type;
import weka.filters.Filter;
import weka.filters.unsupervised.attribute.Normalize;
import weka.filters.unsupervised.attribute.ReplaceMissingValues;
import weka.filters.unsupervised.attribute.Standardize;

import java.util.Enumeration;
import java.util.Vector;

/**
 <!-- globalinfo-start -->
 * Re-implement the Diverse Density algorithm, changes the testing procedure.<br/>
 * <br/>
 * Oded Maron (1998). Learning from ambiguity.<br/>
 * <br/>
 * O. Maron, T. Lozano-Perez (1998). A Framework for Multiple Instance Learning. Neural Information Processing Systems. 10.
 * <p/>
 <!-- globalinfo-end -->
 * 
 <!-- technical-bibtex-start -->
 * BibTeX:
 * <pre>
 * &#64;phdthesis{Maron1998,
 *    author = {Oded Maron},
 *    school = {Massachusetts Institute of Technology},
 *    title = {Learning from ambiguity},
 *    year = {1998}
 * }
 * 
 * &#64;article{Maron1998,
 *    author = {O. Maron and T. Lozano-Perez},
 *    journal = {Neural Information Processing Systems},
 *    title = {A Framework for Multiple Instance Learning},
 *    volume = {10},
 *    year = {1998}
 * }
 * </pre>
 * <p/>
 <!-- technical-bibtex-end -->
 * 
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -D
 *  Turn on debugging output.</pre>
 * 
 * <pre> -N &lt;num&gt;
 *  Whether to 0=normalize/1=standardize/2=neither.
 *  (default 1=standardize)</pre>
 * 
 <!-- options-end -->
 *
 * @author Eibe Frank ([e-mail])
 * @author Xin Xu ([e-mail])
 * @version $Revision: 1.3 $ 
 */
public class MIDD 
  extends Classifier 
  implements OptionHandler, MultiInstanceCapabilitiesHandler,
             TechnicalInformationHandler {

  /** for serialization */
  static final long serialVersionUID = 4263507733600536168L;
  
  /** The index of the class attribute */
  protected int m_ClassIndex;

  protected double[] m_Par;

  /** The number of the class labels */
  protected int m_NumClasses;

  /** Class labels for each bag */
  protected int[] m_Classes;

  /** MI data */ 
  protected double[][][] m_Data;

  /** All attribute names */
  protected Instances m_Attributes;

  /** The filter used to standardize/normalize all values. */
  protected Filter m_Filter = null;

  /** Whether to normalize/standardize/neither, default:standardize */
  protected int m_filterType = FILTER_STANDARDIZE;

  /** Normalize training data */
  public static final int FILTER_NORMALIZE = 0;
  /** Standardize training data */
  public static final int FILTER_STANDARDIZE = 1;
  /** No normalization/standardization */
  public static final int FILTER_NONE = 2;
  /** The filter to apply to the training data */
  public static final Tag [] TAGS_FILTER = {
    new Tag(FILTER_NORMALIZE, "Normalize training data"),
    new Tag(FILTER_STANDARDIZE, "Standardize training data"),
    new Tag(FILTER_NONE, "No normalization/standardization"),
  };

  /** The filter used to get rid of missing values. */
  protected ReplaceMissingValues m_Missing = new ReplaceMissingValues();

  /**
   * Returns a string describing this filter
   *
   * @return a description of the filter suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return 
        "Re-implement the Diverse Density algorithm, changes the testing "
      + "procedure.\n\n"
      + getTechnicalInformation().toString();
  }

  /**
   * Returns an instance of a TechnicalInformation object, containing 
   * detailed information about the technical background of this class,
   * e.g., paper reference or book this class is based on.
   * 
   * @return the technical information about this class
   */
  public TechnicalInformation getTechnicalInformation() {
    TechnicalInformation 	result;
    TechnicalInformation 	additional;
    
    result = new TechnicalInformation(Type.PHDTHESIS);
    result.setValue(Field.AUTHOR, "Oded Maron");
    result.setValue(Field.YEAR, "1998");
    result.setValue(Field.TITLE, "Learning from ambiguity");
    result.setValue(Field.SCHOOL, "Massachusetts Institute of Technology");
    
    additional = result.add(Type.ARTICLE);
    additional.setValue(Field.AUTHOR, "O. Maron and T. Lozano-Perez");
    additional.setValue(Field.YEAR, "1998");
    additional.setValue(Field.TITLE, "A Framework for Multiple Instance Learning");
    additional.setValue(Field.JOURNAL, "Neural Information Processing Systems");
    additional.setValue(Field.VOLUME, "10");
    
    return result;
  }

  /**
   * Returns an enumeration describing the available options
   *
   * @return an enumeration of all the available options
   */
  public Enumeration listOptions() {
    Vector result = new Vector();

    result.addElement(new Option(
          "\tTurn on debugging output.",
          "D", 0, "-D"));

    result.addElement(new Option(
          "\tWhether to 0=normalize/1=standardize/2=neither.\n"
          + "\t(default 1=standardize)",
          "N", 1, "-N <num>"));

    return result.elements();
  }

  /**
   * Parses a given list of options. <p/>
   *     
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <pre> -D
   *  Turn on debugging output.</pre>
   * 
   * <pre> -N &lt;num&gt;
   *  Whether to 0=normalize/1=standardize/2=neither.
   *  (default 1=standardize)</pre>
   * 
   <!-- options-end -->
   *
   * @param options the list of options as an array of strings
   * @throws Exception if an option is not supported
   */
  public void setOptions(String[] options) throws Exception {
    setDebug(Utils.getFlag('D', options));

    String nString = Utils.getOption('N', options);
    if (nString.length() != 0) {
      setFilterType(new SelectedTag(Integer.parseInt(nString), TAGS_FILTER));
    } else {
      setFilterType(new SelectedTag(FILTER_STANDARDIZE, TAGS_FILTER));
    }     
  }

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

    if (getDebug())
      result.add("-D");
    
    result.add("-N");
    result.add("" + m_filterType);

    return (String[]) result.toArray(new String[result.size()]);
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String filterTypeTipText() {
    return "The filter type for transforming the training data.";
  }

  /**
   * Gets how the training data will be transformed. Will be one of
   * FILTER_NORMALIZE, FILTER_STANDARDIZE, FILTER_NONE.
   *
   * @return the filtering mode
   */
  public SelectedTag getFilterType() {
    return new SelectedTag(m_filterType, TAGS_FILTER);
  }

  /**
   * Sets how the training data will be transformed. Should be one of
   * FILTER_NORMALIZE, FILTER_STANDARDIZE, FILTER_NONE.
   *
   * @param newType the new filtering mode
   */
  public void setFilterType(SelectedTag newType) {

    if (newType.getTags() == TAGS_FILTER) {
      m_filterType = newType.getSelectedTag().getID();
    }
  }

  private class OptEng 
    extends Optimization {

    /** 
     * Evaluate objective function
     * @param x the current values of variables
     * @return the value of the objective function 
     */
    protected double objectiveFunction(double[] x){
      double nll = 0; // -LogLikelihood
      for(int i=0; i<m_Classes.length; i++){ // ith bag
        int nI = m_Data[i][0].length; // numInstances in ith bag
        double bag = 0.0;  // NLL of pos bag

        for(int j=0; j<nI; j++){
          double ins=0.0;
          for(int k=0; k<m_Data[i].length; k++)
            ins += (m_Data[i][k][j]-x[k*2])*(m_Data[i][k][j]-x[k*2])*
              x[k*2+1]*x[k*2+1];
          ins = Math.exp(-ins);
          ins = 1.0-ins;

          if(m_Classes[i] == 1)
            bag += Math.log(ins);
          else{
            if(ins<=m_Zero) ins=m_Zero;
            nll -= Math.log(ins);
          }   
        }		

        if(m_Classes[i] == 1){
          bag = 1.0 - Math.exp(bag);
          if(bag<=m_Zero) bag=m_Zero;
          nll -= Math.log(bag);
        }