paceregression.java

wekaUT是 university texas austin 开发的基于weka的半指导学习(semi supervised learning)的分类器

    double[][] result = new double[numInstances]
    [numAttributes];
    for (int i = 0; i < numInstances; i++) {
      Instance inst = data.instance(i);
      
      result[i][0] = 1.0;

      // the class value (lies on index middle) is left out
      for (int j = 0; j < middle; j++) {
	result[i][j + 1] = inst.value(j);
      }
      for (int j = middle + 1; j < numAttributes; j++) {
	result[i][j] = inst.value(j);
      }
    }
    return result;
  }


  /**
   * Classifies the given instance using the linear regression function.
   *
   * @param instance the test instance
   * @return the classification
   * @exception Exception if classification can't be done successfully
   */
  public double classifyInstance(Instance instance) throws Exception {
    
    if (m_Coefficients == null) {
      throw new Exception("Pace Regression: No model built yet.");
    }
    
    // check for missing data and throw exception if some are found
    if (checkForMissing(instance, m_Model)) {
      throw new NoSupportForMissingValuesException("Can't handle missing values!");
    }

    // Calculate the dependent variable from the regression model
    return regressionPrediction(instance,
				m_Coefficients);
  }

  /**
   * Outputs the linear regression model as a string.
   */
  public String toString() {

    if (m_Coefficients == null) {
      return "Pace Regression: No model built yet.";
    }
    //    try {
    StringBuffer text = new StringBuffer();
    
    text.append("\nPace Regression Model\n\n");
    
    text.append(m_Model.classAttribute().name()+" =\n\n");
    int index = 0;	  
    
    text.append(Utils.doubleToString(m_Coefficients[0],
				     12, 4) );
    
    for (int i = 1; i < m_Coefficients.length; i++) {
      
      // jump over the class attribute
      if (index == m_ClassIndex) index++;
      
      if (m_Coefficients[i] != 0.0) {
	// output a coefficient if unequal zero
	text.append(" +\n");
	text.append(Utils.doubleToString(m_Coefficients[i], 12, 4)
		    + " * ");
	text.append(m_Model.attribute(index).name());
      }
      index ++;
    }
    
    return text.toString();
  }
  
  /**
   * Returns an enumeration describing the available options.
   *
   * @return an enumeration of all the available options.
   */
  public Enumeration listOptions() {
    
    Vector newVector = new Vector(2);
    newVector.addElement(new Option("\tProduce debugging output.\n"
				    + "\t(default no debugging output)",
				    "D", 0, "-D"));
    newVector.addElement(new Option("\tThe estimator can be one of the following:\n" + 
				    "\t\teb\tEmpirical Bayes(default)\n" +
				    "\t\tnested\tOptimal nested model\n" + 
				    "\t\tsubset\tOptimal subset\n" +
				    "\t\tpace2\tPACE2\n" +
				    "\t\tpace4\tPACE4\n" +
				    "\t\tpace6\tPACE6\n\n" + 
				    "\t\tols\tOrdinary least squares\n" +  
				    "\t\taic\tAIC\n" +  
				    "\t\tbic\tBIC\n" +  
				    "\t\tric\tRIC\n" +  
				    "\t\tolsc\tOLSC", 
				    "E", 0, "-E <estimator>"));
    newVector.addElement(new Option("\tThreshold value for the OLSC estimator",
				    "S", 0, "-S <threshold value>"));
    return newVector.elements();
  }

  /**
   * Parses a given list of options. <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 {
    
    setDebug(Utils.getFlag('D', options));

    String estimator = Utils.getOption('E', options);
    if ( estimator.equals("ols") ) paceEstimator = olsEstimator;
    else if ( estimator.equals("olsc") ) paceEstimator = olscEstimator;
    else if( estimator.equals("eb") || estimator.equals("") ) 
      paceEstimator = ebEstimator;
    else if ( estimator.equals("nested") ) paceEstimator = nestedEstimator;
    else if ( estimator.equals("subset") ) paceEstimator = subsetEstimator;
    else if ( estimator.equals("pace2") ) paceEstimator = pace2Estimator; 
    else if ( estimator.equals("pace4") ) paceEstimator = pace4Estimator;
    else if ( estimator.equals("pace6") ) paceEstimator = pace6Estimator;
    else if ( estimator.equals("aic") ) paceEstimator = aicEstimator;
    else if ( estimator.equals("bic") ) paceEstimator = bicEstimator;
    else if ( estimator.equals("ric") ) paceEstimator = ricEstimator;
    else throw new WekaException("unknown estimator " + estimator + 
				 " for -E option" );

    String string = Utils.getOption('S', options);
    if( ! string.equals("") ) olscThreshold = Double.parseDouble( string );
    
  }

  /**
   * Returns the coefficients for this linear model.
   */
  public double[] coefficients() {

    double[] coefficients = new double[m_Coefficients.length];
    for (int i = 0; i < coefficients.length; i++) {
      coefficients[i] = m_Coefficients[i];
    }
    return coefficients;
  }

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

    String [] options = new String [6];
    int current = 0;

    if (getDebug()) {
      options[current++] = "-D";
    }

    options[current++] = "-E";
    switch (paceEstimator) {
    case olsEstimator: options[current++] = "ols";
      break;
    case olscEstimator: options[current++] = "olsc";
      options[current++] = "-S";
      options[current++] = "" + olscThreshold;
      break;
    case ebEstimator: options[current++] = "eb";
      break;
    case nestedEstimator: options[current++] = "nested";
      break;
    case subsetEstimator: options[current++] = "subset";
      break;
    case pace2Estimator: options[current++] = "pace2";
      break; 
    case pace4Estimator: options[current++] = "pace4";
      break;
    case pace6Estimator: options[current++] = "pace6";
      break;
    case aicEstimator: options[current++] = "aic";
      break;
    case bicEstimator: options[current++] = "bic";
      break;
    case ricEstimator: options[current++] = "ric";
      break;
    }

    while (current < options.length) {
      options[current++] = "";
    }
    return options;
  }

  
  /**
   * Get the number of coefficients used in the model
   *
   * @return the number of coefficients
   */
  public int numParameters()
  {
    return m_Coefficients.length-1;
  }

  /**
   * Controls whether debugging output will be printed
   *
   * @param debug true if debugging output should be printed
   */
  public void setDebug(boolean debug) {

    m_Debug = debug;
  }

  /**
   * Controls whether debugging output will be printed
   *
   * @param debug true if debugging output should be printed
   */
  public boolean getDebug() {

    return m_Debug;
  }

  /**
   * Gets the estimator
   *
   * @return the estimator
   */
  public SelectedTag getEstimator() {

    return new SelectedTag(paceEstimator, TAGS_ESTIMATOR);
  }
  
  /**
   * Sets the estimator.
   *
   * @param estimator the new estimator
   */
  public void setEstimator(SelectedTag estimator) {
    
    if (estimator.getTags() == TAGS_ESTIMATOR) {
      paceEstimator = estimator.getSelectedTag().getID();
    }
  }

  /**
   * Set threshold for the olsc estimator
   *
   * @param threshold the threshold for the olsc estimator
   */
  public void setThreshold(double newThreshold) {

    olscThreshold = newThreshold;
  }

  /**
   * Gets the threshold for olsc estimator
   *
   * @return the threshold
   */
  public double getThreshold() {

    return olscThreshold;
  }


  /**
   * Calculate the dependent value for a given instance for a
   * given regression model.
   *
   * @param transformedInstance the input instance
   * @param selectedAttributes an array of flags indicating which 
   * attributes are included in the regression model
   * @param coefficients an array of coefficients for the regression
   * model
   * @return the regression value for the instance.
   * @exception Exception if the class attribute of the input instance
   * is not assigned
   */
  private double regressionPrediction(Instance transformedInstance,
				      double [] coefficients) 
    throws Exception {

    int column = 0;
    double result = coefficients[column];
    for (int j = 0; j < transformedInstance.numAttributes(); j++) {
      if (m_ClassIndex != j) {
	column++;
	result += coefficients[column] * transformedInstance.value(j);
      }
    }
    
    return result;
  }

  /**
   * Generates a linear regression function predictor.
   *
   * @param String the options
   */
  public static void main(String argv[]) {
    
    Classifier scheme;
    try {
      scheme = new PaceRegression();
      System.out.println(Evaluation.evaluateModel(scheme, argv));
    } catch (Exception e) {
      e.printStackTrace();
      // System.out.println(e.getMessage());
    }
  }
}