averagingresultproducer.java

Java 编写的多种数据挖掘算法 包括聚类、分类、预处理等

    if ((m_ResultProducer != null) &&
	(m_ResultProducer instanceof OptionHandler)) {
      newVector.addElement(new Option(
	     "",
	     "", 0, "\nOptions specific to result producer "
	     + m_ResultProducer.getClass().getName() + ":"));
      Enumeration enu = ((OptionHandler)m_ResultProducer).listOptions();
      while (enu.hasMoreElements()) {
	newVector.addElement(enu.nextElement());
      }
    }
    return newVector.elements();
  }

  /**
   * Parses a given list of options. <p/>
   *
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <pre> -F &lt;field name&gt;
   *  The name of the field to average over.
   *  (default "Fold")</pre>
   * 
   * <pre> -X &lt;num results&gt;
   *  The number of results expected per average.
   *  (default 10)</pre>
   * 
   * <pre> -S
   *  Calculate standard deviations.
   *  (default only averages)</pre>
   * 
   * <pre> -W &lt;class name&gt;
   *  The full class name of a ResultProducer.
   *  eg: weka.experiment.CrossValidationResultProducer</pre>
   * 
   * <pre> 
   * Options specific to result producer weka.experiment.CrossValidationResultProducer:
   * </pre>
   * 
   * <pre> -X &lt;number of folds&gt;
   *  The number of folds to use for the cross-validation.
   *  (default 10)</pre>
   * 
   * <pre> -D
   * Save raw split evaluator output.</pre>
   * 
   * <pre> -O &lt;file/directory name/path&gt;
   *  The filename where raw output will be stored.
   *  If a directory name is specified then then individual
   *  outputs will be gzipped, otherwise all output will be
   *  zipped to the named file. Use in conjuction with -D. (default splitEvalutorOut.zip)</pre>
   * 
   * <pre> -W &lt;class name&gt;
   *  The full class name of a SplitEvaluator.
   *  eg: weka.experiment.ClassifierSplitEvaluator</pre>
   * 
   * <pre> 
   * Options specific to split evaluator weka.experiment.ClassifierSplitEvaluator:
   * </pre>
   * 
   * <pre> -W &lt;class name&gt;
   *  The full class name of the classifier.
   *  eg: weka.classifiers.bayes.NaiveBayes</pre>
   * 
   * <pre> -C &lt;index&gt;
   *  The index of the class for which IR statistics
   *  are to be output. (default 1)</pre>
   * 
   * <pre> -I &lt;index&gt;
   *  The index of an attribute to output in the
   *  results. This attribute should identify an
   *  instance in order to know which instances are
   *  in the test set of a cross validation. if 0
   *  no output (default 0).</pre>
   * 
   * <pre> -P
   *  Add target and prediction columns to the result
   *  for each fold.</pre>
   * 
   * <pre> 
   * Options specific to classifier weka.classifiers.rules.ZeroR:
   * </pre>
   * 
   * <pre> -D
   *  If set, classifier is run in debug mode and
   *  may output additional info to the console</pre>
   * 
   <!-- options-end -->
   *
   * All options after -- will be passed to the result producer.
   *
   * @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 keyFieldName = Utils.getOption('F', options);
    if (keyFieldName.length() != 0) {
      setKeyFieldName(keyFieldName);
    } else {
      setKeyFieldName(CrossValidationResultProducer.FOLD_FIELD_NAME);
    }

    String numResults = Utils.getOption('X', options);
    if (numResults.length() != 0) {
      setExpectedResultsPerAverage(Integer.parseInt(numResults));
    } else {
      setExpectedResultsPerAverage(10);
    }

    setCalculateStdDevs(Utils.getFlag('S', options));
    
    String rpName = Utils.getOption('W', options);
    if (rpName.length() == 0) {
      throw new Exception("A ResultProducer must be specified with"
			  + " the -W option.");
    }
    // Do it first without options, so if an exception is thrown during
    // the option setting, listOptions will contain options for the actual
    // RP.
    setResultProducer((ResultProducer)Utils.forName(
		      ResultProducer.class,
		      rpName,
		      null));
    if (getResultProducer() instanceof OptionHandler) {
      ((OptionHandler) getResultProducer())
	.setOptions(Utils.partitionOptions(options));
    }
  }

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

    String [] seOptions = new String [0];
    if ((m_ResultProducer != null) && 
	(m_ResultProducer instanceof OptionHandler)) {
      seOptions = ((OptionHandler)m_ResultProducer).getOptions();
    }
    
    String [] options = new String [seOptions.length + 8];
    int current = 0;

    options[current++] = "-F";
    options[current++] = "" + getKeyFieldName();
    options[current++] = "-X";
    options[current++] = "" + getExpectedResultsPerAverage();
    if (getCalculateStdDevs()) {
      options[current++] = "-S";
    }
    if (getResultProducer() != null) {
      options[current++] = "-W";
      options[current++] = getResultProducer().getClass().getName();
    }
    options[current++] = "--";

    System.arraycopy(seOptions, 0, options, current, 
		     seOptions.length);
    current += seOptions.length;
    while (current < options.length) {
      options[current++] = "";
    }
    return options;
  }

  /**
   * Set a list of method names for additional measures to look for
   * in SplitEvaluators. This could contain many measures (of which only a
   * subset may be produceable by the current resultProducer) if an experiment
   * is the type that iterates over a set of properties.
   * @param additionalMeasures an array of measure names, null if none
   */
  public void setAdditionalMeasures(String [] additionalMeasures) {
    m_AdditionalMeasures = additionalMeasures;

    if (m_ResultProducer != null) {
      System.err.println("AveragingResultProducer: setting additional "
			 +"measures for "
			 +"ResultProducer");
      m_ResultProducer.setAdditionalMeasures(m_AdditionalMeasures);
    }
  }

  /**
   * Returns an enumeration of any additional measure names that might be
   * in the result producer
   * @return an enumeration of the measure names
   */
  public Enumeration enumerateMeasures() {
    Vector newVector = new Vector();
    if (m_ResultProducer instanceof AdditionalMeasureProducer) {
      Enumeration en = ((AdditionalMeasureProducer)m_ResultProducer).
	enumerateMeasures();
      while (en.hasMoreElements()) {
	String mname = (String)en.nextElement();
	newVector.addElement(mname);
      }
    }
    return newVector.elements();
  }

  /**
   * Returns the value of the named measure
   * @param additionalMeasureName the name of the measure to query for its value
   * @return the value of the named measure
   * @throws IllegalArgumentException if the named measure is not supported
   */
  public double getMeasure(String additionalMeasureName) {
    if (m_ResultProducer instanceof AdditionalMeasureProducer) {
      return ((AdditionalMeasureProducer)m_ResultProducer).
	getMeasure(additionalMeasureName);
    } else {
      throw new IllegalArgumentException("AveragingResultProducer: "
			  +"Can't return value for : "+additionalMeasureName
			  +". "+m_ResultProducer.getClass().getName()+" "
			  +"is not an AdditionalMeasureProducer");
    }
  }

  /**
   * Sets the dataset that results will be obtained for.
   *
   * @param instances a value of type 'Instances'.
   */
  public void setInstances(Instances instances) {
    
    m_Instances = instances;
  }

  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String calculateStdDevsTipText() {
    return "Record standard deviations for each run.";
  }

  /**
   * Get the value of CalculateStdDevs.
   *
   * @return Value of CalculateStdDevs.
   */
  public boolean getCalculateStdDevs() {
    
    return m_CalculateStdDevs;
  }
  
  /**
   * Set the value of CalculateStdDevs.
   *
   * @param newCalculateStdDevs Value to assign to CalculateStdDevs.
   */
  public void setCalculateStdDevs(boolean newCalculateStdDevs) {
    
    m_CalculateStdDevs = newCalculateStdDevs;
  }

  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String expectedResultsPerAverageTipText() {
    return "Set the expected number of results to average per run. "
      +"For example if a CrossValidationResultProducer is being used "
      +"(with the number of folds set to 10), then the expected number "
      +"of results per run is 10.";
  }

  /**
   * Get the value of ExpectedResultsPerAverage.
   *
   * @return Value of ExpectedResultsPerAverage.
   */
  public int getExpectedResultsPerAverage() {
    
    return m_ExpectedResultsPerAverage;
  }
  
  /**
   * Set the value of ExpectedResultsPerAverage.
   *
   * @param newExpectedResultsPerAverage Value to assign to
   * ExpectedResultsPerAverage.
   */
  public void setExpectedResultsPerAverage(int newExpectedResultsPerAverage) {
    
    m_ExpectedResultsPerAverage = newExpectedResultsPerAverage;
  }

  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String keyFieldNameTipText() {
    return "Set the field name that will be unique for a run.";
  }

  /**
   * Get the value of KeyFieldName.
   *
   * @return Value of KeyFieldName.
   */
  public String getKeyFieldName() {
    
    return m_KeyFieldName;
  }
  
  /**
   * Set the value of KeyFieldName.
   *
   * @param newKeyFieldName Value to assign to KeyFieldName.
   */
  public void setKeyFieldName(String newKeyFieldName) {
    
    m_KeyFieldName = newKeyFieldName;
    m_CountFieldName = "Num_" + m_KeyFieldName;
    findKeyIndex();
  }
  
  /**
   * Sets the object to send results of each run to.
   *
   * @param listener a value of type 'ResultListener'
   */
  public void setResultListener(ResultListener listener) {

    m_ResultListener = listener;
  }
  
  /**
   * Returns the tip text for this property
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String resultProducerTipText() {
    return "Set the resultProducer for which results are to be averaged.";
  }

  /**
   * Get the ResultProducer.
   *
   * @return the ResultProducer.
   */
  public ResultProducer getResultProducer() {
    
    return m_ResultProducer;
  }
  
  /**
   * Set the ResultProducer.
   *
   * @param newResultProducer new ResultProducer to use.
   */
  public void setResultProducer(ResultProducer newResultProducer) {

    m_ResultProducer = newResultProducer;
    m_ResultProducer.setResultListener(this);
    findKeyIndex();
  }

  /**
   * Gets a text descrption of the result producer.
   *
   * @return a text description of the result producer.
   */
  public String toString() {

    String result = "AveragingResultProducer: ";
    result += getCompatibilityState();
    if (m_Instances == null) {
      result += ": <null Instances>";
    } else {
      result += ": " + Utils.backQuoteChars(m_Instances.relationName());
    }
    return result;
  }

} // AveragingResultProducer