attributeselection.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

  /**
   * Gets the current settings for the attribute selection (search, evaluator)
   * etc.
   *
   * @return an array of strings suitable for passing to setOptions()
   */
  public String [] getOptions() {
    String [] EvaluatorOptions = new String[0];
    String [] SearchOptions = new String[0];
    int current = 0;

    if (m_ASEvaluator instanceof OptionHandler) {
      EvaluatorOptions = ((OptionHandler)m_ASEvaluator).getOptions();
    }

    if (m_ASSearch instanceof OptionHandler) {
      SearchOptions = ((OptionHandler)m_ASSearch).getOptions();
    }

    String [] setOptions = new String [10];
    setOptions[current++]="-E";
    setOptions[current++]= getEvaluator().getClass().getName()
      +" "+Utils.joinOptions(EvaluatorOptions);

    setOptions[current++]="-S";
    setOptions[current++]=getSearch().getClass().getName() 
      + " "+Utils.joinOptions(SearchOptions);

    while (current < setOptions.length) {
      setOptions[current++] = "";
    }
    
    return setOptions;
  }
  
  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String evaluatorTipText() {

    return "Determines how attributes/attribute subsets are evaluated.";
  }

  /**
   * set attribute/subset evaluator
   * 
   * @param evaluator the evaluator to use
   */
  public void setEvaluator(ASEvaluation evaluator) {
    m_ASEvaluator = evaluator;
  }
  
  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String searchTipText() {

    return "Determines the search method.";
  }

  /**
   * Set search class
   * 
   * @param search the search class to use
   */
  public void setSearch(ASSearch search) {
    m_ASSearch = search;
  }

  /**
   * Get the name of the attribute/subset evaluator
   *
   * @return the name of the attribute/subset evaluator as a string
   */
  public ASEvaluation getEvaluator() {
    
      return m_ASEvaluator;
  }

  /**
   * Get the name of the search method
   *
   * @return the name of the search method as a string
   */
  public ASSearch getSearch() {
    
      return m_ASSearch;
  }

  /** 
   * Returns the Capabilities of this filter.
   *
   * @return            the capabilities of this object
   * @see               Capabilities
   */
  public Capabilities getCapabilities() {
    Capabilities	result;
    
    if (m_ASEvaluator == null) {
      result = super.getCapabilities();
    }
    else {
      result = m_ASEvaluator.getCapabilities();
      // class index will be set if necessary, so we always allow the dataset
      // to have no class attribute set. see the following method:
      //   weka.attributeSelection.AttributeSelection.SelectAttributes(Instances)
      result.enable(Capability.NO_CLASS);
    }
    
    result.setMinimumNumberInstances(0);
    
    return result;
  }

  /**
   * Input an instance for filtering. Ordinarily the instance is processed
   * and made available for output immediately. Some filters require all
   * instances be read before producing output.
   *
   * @param instance the input instance
   * @return true if the filtered instance may now be
   * collected with output().
   * @throws IllegalStateException if no input format has been defined.
   * @throws Exception if the input instance was not of the correct format 
   * or if there was a problem with the filtering.
   */
  public boolean input(Instance instance) throws Exception {
    
    if (getInputFormat() == null) {
      throw new IllegalStateException("No input instance format defined");
    }

    if (m_NewBatch) {
      resetQueue();
      m_NewBatch = false;
    }

    if (isOutputFormatDefined()) {
      convertInstance(instance);
      return true;
    }

    bufferInput(instance);
    return false;
  }

  /**
   * Signify that this batch of input to the filter is finished. If the filter
   * requires all instances prior to filtering, output() may now be called
   * to retrieve the filtered instances.
   *
   * @return true if there are instances pending output.
   * @throws IllegalStateException if no input structure has been defined.
   * @throws Exception if there is a problem during the attribute selection.
   */
  public boolean batchFinished() throws Exception {
    
    if (getInputFormat() == null) {
      throw new IllegalStateException("No input instance format defined");
    }

    if (!isOutputFormatDefined()) {
      m_trainSelector.setEvaluator(m_ASEvaluator);
      m_trainSelector.setSearch(m_ASSearch);
      m_trainSelector.SelectAttributes(getInputFormat());
      //      System.out.println(m_trainSelector.toResultsString());

      m_SelectedAttributes = m_trainSelector.selectedAttributes();
      if (m_SelectedAttributes == null) {
	throw new Exception("No selected attributes\n");
      }
     
      setOutputFormat();
      
      // Convert pending input instances
      for (int i = 0; i < getInputFormat().numInstances(); i++) {
	convertInstance(getInputFormat().instance(i));
      }
      flushInput();
    }
    
    m_NewBatch = true;
    return (numPendingOutput() != 0);
  }

  /**
   * Set the output format. Takes the currently defined attribute set 
   * m_InputFormat and calls setOutputFormat(Instances) appropriately.
   * 
   * @throws Exception if something goes wrong
   */
  protected void setOutputFormat() throws Exception {
    Instances informat;

    if (m_SelectedAttributes == null) {
      setOutputFormat(null);
      return;
    }

    FastVector attributes = new FastVector(m_SelectedAttributes.length);

    int i;
    if (m_ASEvaluator instanceof AttributeTransformer) {
      informat = ((AttributeTransformer)m_ASEvaluator).transformedData();
    } else {
      informat = getInputFormat();
    }

    for (i=0;i < m_SelectedAttributes.length;i++) {
      attributes.
	addElement(informat.attribute(m_SelectedAttributes[i]).copy());
    }

    Instances outputFormat = 
      new Instances(getInputFormat().relationName(), attributes, 0);


    if (!(m_ASEvaluator instanceof UnsupervisedSubsetEvaluator) &&
	!(m_ASEvaluator instanceof UnsupervisedAttributeEvaluator)) {
      outputFormat.setClassIndex(m_SelectedAttributes.length - 1);
    }
    
    setOutputFormat(outputFormat);  
  }

  /**
   * Convert a single instance over. Selected attributes only are transfered.
   * The converted instance is added to the end of
   * the output queue.
   *
   * @param instance the instance to convert
   * @throws Exception if something goes wrong
   */
  protected void convertInstance(Instance instance) throws Exception {
    double[] newVals = new double[getOutputFormat().numAttributes()];

    if (m_ASEvaluator instanceof AttributeTransformer) {
      Instance tempInstance = ((AttributeTransformer)m_ASEvaluator).
	convertInstance(instance);
      for (int i = 0; i < m_SelectedAttributes.length; i++) {
	int current = m_SelectedAttributes[i];
	newVals[i] = tempInstance.value(current);
      }
    } else {
      for (int i = 0; i < m_SelectedAttributes.length; i++) {
	int current = m_SelectedAttributes[i];
	newVals[i] = instance.value(current);
      }
    }
    if (instance instanceof SparseInstance) {
      push(new SparseInstance(instance.weight(), newVals));
    } else {
      push(new Instance(instance.weight(), newVals));
    }
  }

  /**
   * set options to their default values
   */
  protected void resetOptions() {

    m_trainSelector = new weka.attributeSelection.AttributeSelection();
    setEvaluator(new CfsSubsetEval());
    setSearch(new BestFirst());
    m_SelectedAttributes = null;
    m_FilterOptions = null;
  }

  /**
   * 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 AttributeSelection(), argv);
  }
}