removemisclassified.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

	      + "\tIf < 0 will use any current set class or default to the last attribute.",
	      "C", 1, "-C <class index>"));
    newVector.addElement(new Option(
	      "\tThe number of folds to use for cross-validation cleansing.\n"
	      +"\t(<2 = no cross-validation - default).",
	      "F", 1, "-F <number of folds>"));
    newVector.addElement(new Option(
	      "\tThreshold for the max error when predicting numeric class.\n"
	      +"\t(Value should be >= 0, default = 0.1).",
	      "T", 1, "-T <threshold>"));
    newVector.addElement(new Option(
	      "\tThe maximum number of cleansing iterations to perform.\n"
	      +"\t(<1 = until fully cleansed - default)",
	      "I", 1,"-I"));
    newVector.addElement(new Option(
	      "\tInvert the match so that correctly classified instances are discarded.\n",
	      "V", 0,"-V"));
    
    return newVector.elements();
  }


  /**
   * Parses a given list of options. <p/>
   * 
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <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> -C &lt;class index&gt;
   *  Attribute on which misclassifications are based.
   *  If &lt; 0 will use any current set class or default to the last attribute.</pre>
   * 
   * <pre> -F &lt;number of folds&gt;
   *  The number of folds to use for cross-validation cleansing.
   *  (&lt;2 = no cross-validation - default).</pre>
   * 
   * <pre> -T &lt;threshold&gt;
   *  Threshold for the max error when predicting numeric class.
   *  (Value should be &gt;= 0, default = 0.1).</pre>
   * 
   * <pre> -I
   *  The maximum number of cleansing iterations to perform.
   *  (&lt;1 = until fully cleansed - default)</pre>
   * 
   * <pre> -V
   *  Invert the match so that correctly classified instances are discarded.
   * </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 {

    String classifierString = Utils.getOption('W', options);
    if (classifierString.length() == 0)
      classifierString = weka.classifiers.rules.ZeroR.class.getName();
    String[] classifierSpec = Utils.splitOptions(classifierString);
    if (classifierSpec.length == 0) {
      throw new Exception("Invalid classifier specification string");
    }
    String classifierName = classifierSpec[0];
    classifierSpec[0] = "";
    setClassifier(Classifier.forName(classifierName, classifierSpec));

    String cString = Utils.getOption('C', options);
    if (cString.length() != 0) {
      setClassIndex((new Double(cString)).intValue());
    } else {
      setClassIndex(-1);
    }

    String fString = Utils.getOption('F', options);
    if (fString.length() != 0) {
      setNumFolds((new Double(fString)).intValue());
    } else {
      setNumFolds(0);
    }

    String tString = Utils.getOption('T', options);
    if (tString.length() != 0) {
      setThreshold((new Double(tString)).doubleValue());
    } else {
      setThreshold(0.1);
    }

    String iString = Utils.getOption('I', options);
    if (iString.length() != 0) {
      setMaxIterations((new Double(iString)).intValue());
    } else {
      setMaxIterations(0);
    }
    
    if (Utils.getFlag('V', options)) {
      setInvert(true);
    } else {
      setInvert(false);
    }
        
    Utils.checkForRemainingOptions(options);

  }

  /**
   * Gets the current settings of the filter.
   *
   * @return an array of strings suitable for passing to setOptions
   */
  public String [] getOptions() {

    String [] options = new String [15];
    int current = 0;

    options[current++] = "-W"; options[current++] = "" + getClassifierSpec();
    options[current++] = "-C"; options[current++] = "" + getClassIndex();
    options[current++] = "-F"; options[current++] = "" + getNumFolds();
    options[current++] = "-T"; options[current++] = "" + getThreshold();
    options[current++] = "-I"; options[current++] = "" + getMaxIterations();
    if (getInvert()) {
      options[current++] = "-V";
    }
    
    while (current < options.length) {
      options[current++] = "";
    }
    return options;
  }

  /**
   * 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 that removes instances which are incorrectly classified. "
      + "Useful for removing outliers.";
  }

  /**
   * 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 upon which to base the misclassifications.";
  }

  /**
   * Sets the classifier to classify instances with.
   *
   * @param classifier The classifier to be used (with its options set).
   */
  public void setClassifier(Classifier classifier) {

    m_cleansingClassifier = classifier;
  }
  
  /**
   * Gets the classifier used by the filter.
   *
   * @return The classifier to be used.
   */
  public Classifier getClassifier() {

    return m_cleansingClassifier;
  }

  /**
   * Gets the classifier specification string, which contains the class name of
   * the classifier and any options to the classifier.
   *
   * @return the classifier string.
   */
  protected String getClassifierSpec() {
    
    Classifier c = getClassifier();
    if (c instanceof OptionHandler) {
      return c.getClass().getName() + " "
	+ Utils.joinOptions(((OptionHandler)c).getOptions());
    }
    return c.getClass().getName();
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String classIndexTipText() {

    return "Index of the class upon which to base the misclassifications. "
      + "If < 0 will use any current set class or default to the last attribute.";
  }

  /**
   * Sets the attribute on which misclassifications are based.
   * If &lt; 0 will use any current set class or default to the last attribute.
   *
   * @param classIndex the class index.
   */
  public void setClassIndex(int classIndex) {
    
    m_classIndex = classIndex;
  }

  /**
   * Gets the attribute on which misclassifications are based.
   *
   * @return the class index.
   */
  public int getClassIndex() {

    return m_classIndex;
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String numFoldsTipText() {

    return "The number of cross-validation folds to use. If < 2 then no cross-validation will be performed.";
  }

  /**
   * Sets the number of cross-validation folds to use
   * - &lt; 2 means no cross-validation.
   *
   * @param numOfFolds the number of folds.
   */
  public void setNumFolds(int numOfFolds) {
    
    m_numOfCrossValidationFolds = numOfFolds;
  }

  /**
   * Gets the number of cross-validation folds used by the filter.
   *
   * @return the number of folds.
   */
  public int getNumFolds() {

    return m_numOfCrossValidationFolds;
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String thresholdTipText() {

    return "Threshold for the max allowable error when predicting a numeric class. Should be >= 0.";
  }

  /**
   * Sets the threshold for the max error when predicting a numeric class.
   * The value should be &gt;= 0.
   *
   * @param threshold the numeric theshold.
   */
  public void setThreshold(double threshold) {
    
    m_numericClassifyThreshold = threshold;
  }

  /**
   * Gets the threshold for the max error when predicting a numeric class.
   *
   * @return the numeric threshold.
   */
  public double getThreshold() {

    return m_numericClassifyThreshold;
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String maxIterationsTipText() {

    return "The maximum number of iterations to perform. < 1 means filter will go until fully cleansed.";
  }

  /**
   * Sets the maximum number of cleansing iterations to perform
   * - &lt; 1 means go until fully cleansed
   *
   * @param iterations the maximum number of iterations.
   */
  public void setMaxIterations(int iterations) {
    
    m_numOfCleansingIterations = iterations;
  }

  /**
   * Gets the maximum number of cleansing iterations performed
   *
   * @return the maximum number of iterations. 
   */
  public int getMaxIterations() {

    return m_numOfCleansingIterations;
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String invertTipText() {

    return "Whether or not to invert the selection. If true, correctly classified instances will be discarded.";
  }

  /**
   * Set whether selection is inverted.
   *
   * @param invert whether or not to invert selection.
   */
  public void setInvert(boolean invert) {
    
    m_invertMatching = invert;
  }

  /**
   * Get whether selection is inverted.
   *
   * @return whether or not selection is inverted.
   */
  public boolean getInvert() {
    
    return m_invertMatching;
  }

  /**
   * Main method for testing this class.
   *
   * @param argv should contain arguments to the filter: use -h for help
   */
  public static void main(String [] argv) {
    runFilter(new RemoveMisclassified(), argv);
  }
}