addclassification.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.
 */

/*
 * AddClassification.java
 * Copyright (C) 2006 University of Waikato, Hamilton, New Zealand
 */

package weka.filters.supervised.attribute;

import weka.classifiers.Classifier;
import weka.core.Attribute;
import weka.core.Capabilities;
import weka.core.FastVector;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.SparseInstance;
import weka.core.Utils;
import weka.filters.SimpleBatchFilter;

import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.ObjectInputStream;
import java.util.Enumeration;
import java.util.Vector;

/**
 <!-- globalinfo-start -->
 * A filter for adding the classification, the class distribution and an error flag to a dataset with a classifier. The classifier is either trained on the data itself or provided as serialized model.
 * <p/>
 <!-- globalinfo-end -->
 * 
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -D
 *  Turns on output of debugging information.</pre>
 * 
 * <pre> -W &lt;classifier specification&gt;
 *  Full class name of classifier to use, followed
 *  by scheme options. eg:
 *   "weka.classifiers.bayes.NaiveBayes -D"
 *  (default: weka.classifiers.rules.ZeroR)</pre>
 * 
 * <pre> -serialized &lt;file&gt;
 *  Instead of training a classifier on the data, one can also provide
 *  a serialized model and use that for tagging the data.</pre>
 * 
 * <pre> -classification
 *  Adds an attribute with the actual classification.
 *  (default: off)</pre>
 * 
 * <pre> -distribution
 *  Adds attributes with the distribution for all classes 
 *  (for numeric classes this will be identical to the attribute 
 *  output with '-classification').
 *  (default: off)</pre>
 * 
 * <pre> -error
 *  Adds an attribute indicating whether the classifier output 
 *  a wrong classification (for numeric classes this is the numeric 
 *  difference).
 *  (default: off)</pre>
 * 
 <!-- options-end -->
 *
 * @author  fracpete (fracpete at waikato dot ac dot nz)
 * @version $Revision: 1.1 $
 */
public class AddClassification
  extends SimpleBatchFilter {

  /** for serialization */
  private static final long serialVersionUID = -1931467132568441909L;

  /** The classifier template used to do the classification */
  protected Classifier m_Classifier = new weka.classifiers.rules.ZeroR();

  /** The file from which to load a serialized classifier */
  protected File m_SerializedClassifierFile = new File(System.getProperty("user.dir"));
  
  /** The actual classifier used to do the classification */
  protected Classifier m_ActualClassifier = null;

  /** whether to output the classification */
  protected boolean m_OutputClassification = false;
  
  /** whether to output the class distribution */
  protected boolean m_OutputDistribution = false;
  
  /** whether to output the error flag */
  protected boolean m_OutputErrorFlag = false;

  /**
   * Returns a string describing this filter
   *
   * @return 		a description of the filter suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return 
        "A filter for adding the classification, the class distribution and "
      + "an error flag to a dataset with a classifier. The classifier is "
      + "either trained on the data itself or provided as serialized model.";
  }

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

    result = new Vector();

    en = super.listOptions();
    while (en.hasMoreElements())
      result.addElement(en.nextElement());

    result.addElement(new Option(
	"\tFull class name of classifier to use, followed\n"
	+ "\tby scheme options. eg:\n"
	+ "\t\t\"weka.classifiers.bayes.NaiveBayes -D\"\n"
	+ "\t(default: weka.classifiers.rules.ZeroR)",
	"W", 1, "-W <classifier specification>"));

    result.addElement(new Option(
	"\tInstead of training a classifier on the data, one can also provide\n"
	+ "\ta serialized model and use that for tagging the data.",
	"serialized", 1, "-serialized <file>"));

    result.addElement(new Option(
	"\tAdds an attribute with the actual classification.\n"
	+ "\t(default: off)",
	"classification", 0, "-classification"));

    result.addElement(new Option(
	"\tAdds attributes with the distribution for all classes \n"
        + "\t(for numeric classes this will be identical to the attribute \n"
        + "\toutput with '-classification').\n"
	+ "\t(default: off)",
	"distribution", 0, "-distribution"));

    result.addElement(new Option(
	"\tAdds an attribute indicating whether the classifier output \n"
        + "\ta wrong classification (for numeric classes this is the numeric \n"
        + "\tdifference).\n"
	+ "\t(default: off)",
	"error", 0, "-error"));

    return result.elements();
  }

  /**
   * Parses the options for this object. <p/>
   *
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <pre> -D
   *  Turns on output of debugging information.</pre>
   * 
   * <pre> -W &lt;classifier specification&gt;
   *  Full class name of classifier to use, followed
   *  by scheme options. eg:
   *   "weka.classifiers.bayes.NaiveBayes -D"
   *  (default: weka.classifiers.rules.ZeroR)</pre>
   * 
   * <pre> -serialized &lt;file&gt;
   *  Instead of training a classifier on the data, one can also provide
   *  a serialized model and use that for tagging the data.</pre>
   * 
   * <pre> -classification
   *  Adds an attribute with the actual classification.
   *  (default: off)</pre>
   * 
   * <pre> -distribution
   *  Adds attributes with the distribution for all classes 
   *  (for numeric classes this will be identical to the attribute 
   *  output with '-classification').
   *  (default: off)</pre>
   * 
   * <pre> -error
   *  Adds an attribute indicating whether the classifier output 
   *  a wrong classification (for numeric classes this is the numeric 
   *  difference).
   *  (default: off)</pre>
   * 
   <!-- options-end -->
   *
   * @param options	the options to use
   * @throws Exception	if setting of options fails
   */
  public void setOptions(String[] options) throws Exception {
    String	tmpStr;
    String[] 	tmpOptions;
    File	file;
    boolean 	serializedModel;

    setOutputClassification(Utils.getFlag("classification", options));
    
    setOutputClassification(Utils.getFlag("distribution", options));

    setOutputClassification(Utils.getFlag("error", options));
    
    serializedModel = false;
    tmpStr = Utils.getOption("serialized", options);
    if (tmpStr.length() != 0) {
      file = new File(tmpStr);
      if (!file.exists())
	throw new FileNotFoundException(
	    "File '" + file.getAbsolutePath() + "' not found!");
      if (file.isDirectory())
	throw new FileNotFoundException(
	    "'" + file.getAbsolutePath() + "' points to a directory not a file!");
      setSerializedClassifierFile(file);
      serializedModel = true;
    }
    else {
      setSerializedClassifierFile(null);
    }
    
    if (!serializedModel) {
      tmpStr = Utils.getOption('W', options);
      if (tmpStr.length() == 0)
	tmpStr = weka.classifiers.rules.ZeroR.class.getName();
      tmpOptions = Utils.splitOptions(tmpStr);
      if (tmpOptions.length == 0)
	throw new Exception("Invalid classifier specification string");
      tmpStr = tmpOptions[0];
      tmpOptions[0] = "";
      setClassifier(Classifier.forName(tmpStr, tmpOptions));
    }

    super.setOptions(options);
  }

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

    result = new Vector();

    options = super.getOptions();
    for (i = 0; i < options.length; i++)
      result.add(options[i]);

    if (getOutputClassification())
      result.add("-classification");

    if (getOutputDistribution())
      result.add("-distribution");

    if (getOutputErrorFlag())
      result.add("-error");

    file = getSerializedClassifierFile();
    if ((file != null) && (!file.isDirectory())) {
      result.add("-serialized");
      result.add(file.getAbsolutePath());
    }
    else {
      result.add("-W");
      result.add(getClassifierSpec());
    }
    
    return (String[]) result.toArray(new String[result.size()]);	  
  }

  /** 
   * Returns the Capabilities of this filter.
   *
   * @return            the capabilities of this object
   * @see               Capabilities
   */
  public Capabilities getCapabilities() {
    Capabilities 	result;
    
    if (getClassifier() == null)
      result = super.getCapabilities();
    else
      result = getClassifier().getCapabilities();
    
    result.setMinimumNumberInstances(0);
    
    return result;
  }

  /**
   * 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 "The classifier to use for classification.";
  }

  /**