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

   * Sets the classifier to classify instances with.
   *
   * @param value 	The classifier to be used (with its options set).
   */
  public void setClassifier(Classifier value) {
    m_Classifier = value;
  }
  
  /**
   * Gets the classifier used by the filter.
   *
   * @return 		The classifier to be used.
   */
  public Classifier getClassifier() {
    return m_Classifier;
  }

  /**
   * 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() {
    String	result;
    Classifier 	c;
    
    c      = getClassifier();
    result = c.getClass().getName();
    if (c instanceof OptionHandler)
      result += " " + Utils.joinOptions(((OptionHandler) c).getOptions());
    
    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 serializedClassifierFileTipText() {
    return "A file containing the serialized model of a trained classifier.";
  }

  /**
   * Gets the file pointing to a serialized, trained classifier. If it is
   * null or pointing to a directory it will not be used.
   * 
   * @return		the file the serialized, trained classifier is located 
   * 			in
   */
  public File getSerializedClassifierFile() {
    return m_SerializedClassifierFile;
  }

  /**
   * Sets the file pointing to a serialized, trained classifier. If the
   * argument is null, doesn't exist or pointing to a directory, then the 
   * value is ignored.
   * 
   * @param value	the file pointing to the serialized, trained classifier
   */
  public void setSerializedClassifierFile(File value) {
    if ((value == null) || (!value.exists()))
      value = new File(System.getProperty("user.dir"));

    m_SerializedClassifierFile = value;
  }
  
  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String outputClassificationTipText() {
    return "Whether to add an attribute with the actual classification.";
  }

  /**
   * Get whether the classifiction of the classifier is output.
   *
   * @return 		true if the classification of the classifier is output.
   */
  public boolean getOutputClassification() {
    return m_OutputClassification;
  }
  
  /**
   * Set whether the classification of the classifier is output.
   *
   * @param value 	whether the classification of the classifier is output.
   */
  public void setOutputClassification(boolean value) {
    m_OutputClassification = value;
  }
  
  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String outputDistributionTipText() {
    return 
        "Whether to add attributes with the distribution for all classes "
      + "(for numeric classes this will be identical to the attribute output "
      + "with 'outputClassification').";
  }

  /**
   * Get whether the classifiction of the classifier is output.
   *
   * @return 		true if the distribution of the classifier is output.
   */
  public boolean getOutputDistribution() {
    return m_OutputDistribution;
  }
  
  /**
   * Set whether the Distribution of the classifier is output.
   *
   * @param value 	whether the distribution of the classifier is output.
   */
  public void setOutputDistribution(boolean value) {
    m_OutputDistribution = value;
  }
  
  /**
   * Returns the tip text for this property
   * 
   * @return 		tip text for this property suitable for
   * 			displaying in the explorer/experimenter gui
   */
  public String outputErrorFlagTipText() {
    return 
        "Whether to add an attribute indicating whether the classifier output "
      + "a wrong classification (for numeric classes this is the numeric "
      + "difference).";
  }

  /**
   * Get whether the classifiction of the classifier is output.
   *
   * @return 		true if the classification of the classifier is output.
   */
  public boolean getOutputErrorFlag() {
    return m_OutputErrorFlag;
  }
  
  /**
   * Set whether the classification of the classifier is output.
   *
   * @param value 	whether the classification of the classifier is output.
   */
  public void setOutputErrorFlag(boolean value) {
    m_OutputErrorFlag = value;
  }

  /**
   * Determines the output format based on the input format and returns 
   * this. In case the output format cannot be returned immediately, i.e.,
   * immediateOutputFormat() returns false, then this method will be called
   * from batchFinished().
   *
   * @param inputFormat     the input format to base the output format on
   * @return                the output format
   * @throws Exception      in case the determination goes wrong
   * @see   #hasImmediateOutputFormat()
   * @see   #batchFinished()
   */
  protected Instances determineOutputFormat(Instances inputFormat)
      throws Exception {
    
    Instances	result;
    FastVector	atts;
    int		i;
    FastVector	values;
    
    // copy old attributes
    atts = new FastVector();
    for (i = 0; i < inputFormat.numAttributes(); i++)
      atts.addElement(inputFormat.attribute(i).copy());
    
    // add new attributes
    // 1. classification?
    if (getOutputClassification()) {
      atts.addElement(inputFormat.classAttribute().copy("classification"));
    }
    
    // 2. distribution?
    if (getOutputDistribution()) {
      if (inputFormat.classAttribute().isNominal()) {
	for (i = 0; i < inputFormat.classAttribute().numValues(); i++) {
	  atts.addElement(new Attribute("distribution_" + inputFormat.classAttribute().value(i)));
	}
      }
      else {
	atts.addElement(new Attribute("distribution"));
      }
    }
    
    // 2. error flag?
    if (getOutputErrorFlag()) {
      if (inputFormat.classAttribute().isNominal()) {
	values = new FastVector();
	values.addElement("no");
	values.addElement("yes");
	atts.addElement(new Attribute("error", values));
      }
      else {
	atts.addElement(new Attribute("error"));
      }
    }
    
    // generate new header
    result = new Instances(inputFormat.relationName(), atts, 0);
    result.setClassIndex(inputFormat.classIndex());
    
    return result;
  }

  /**
   * Processes the given data (may change the provided dataset) and returns
   * the modified version. This method is called in batchFinished().
   *
   * @param instances   the data to process
   * @return            the modified data
   * @throws Exception  in case the processing goes wrong
   * @see               #batchFinished()
   */
  protected Instances process(Instances instances) throws Exception {
    Instances		result;
    double[]		newValues;
    double[]		oldValues;
    int			i;
    int			start;
    int			n;
    Instance		newInstance;
    Instance		oldInstance;
    double[]		distribution;
    File		file;
    ObjectInputStream 	ois;
    
    // load or train classifier
    if (!isFirstBatchDone()) {
      file = getSerializedClassifierFile();
      if (!file.isDirectory()) {
	ois = new ObjectInputStream(new FileInputStream(file));
	m_ActualClassifier = (Classifier) ois.readObject();
	ois.close();
      }
      else {
	m_ActualClassifier = Classifier.makeCopy(m_Classifier);
	m_ActualClassifier.buildClassifier(instances);
      }
    }
    
    result = getOutputFormat();
    
    // traverse all instances
    for (i = 0; i < instances.numInstances(); i++) {
      oldInstance = instances.instance(i);
      oldValues   = oldInstance.toDoubleArray();
      newValues   = new double[result.numAttributes()];
      
      // copy old values
      System.arraycopy(oldValues, 0, newValues, 0, oldValues.length);
      
      // add new values:
      start = oldValues.length;
      
      // 1. classification?
      if (getOutputClassification()) {
	newValues[start] = m_ActualClassifier.classifyInstance(oldInstance);
	start++;
      }
      
      // 2. distribution?
      if (getOutputDistribution()) {
	distribution = m_ActualClassifier.distributionForInstance(oldInstance);
	for (n = 0; n < distribution.length; n++) {
	  newValues[start] = distribution[n];
	  start++;
	}
      }
      
      // 3. error flag?
      if (getOutputErrorFlag()) {
	if (result.classAttribute().isNominal()) {
	  if (oldInstance.classValue() == m_ActualClassifier.classifyInstance(oldInstance))
	    newValues[start] = 0;
	  else
	    newValues[start] = 1;
	}
	else {
	  newValues[start] = m_ActualClassifier.classifyInstance(oldInstance) - oldInstance.classValue();
	}
	start++;
      }
      
      // create new instance
      if (oldInstance instanceof SparseInstance)
	newInstance = new SparseInstance(oldInstance.weight(), newValues);
      else
	newInstance = new Instance(oldInstance.weight(), newValues);
      result.add(newInstance);
    }
    
    return result;
  }

  /**
   * runs the filter with the given arguments
   *
   * @param args      the commandline arguments
   */
  public static void main(String[] args) {
    runFilter(new AddClassification(), args);
  }
}