multiinstancetopropositional.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

/*
 *    This program is free software; you can redistribute it and/or modify
 *    it under the terms of the GNU General Public License as published by
 *    the Free Software Foundation; either version 2 of the License, or
 *    (at your option) any later version.
 *
 *    This program is distributed in the hope that it will be useful,
 *    but WITHOUT ANY WARRANTY; without even the implied warranty of
 *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *    GNU General Public License for more details.
 *
 *    You should have received a copy of the GNU General Public License
 *    along with this program; if not, write to the Free Software
 *    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

/*
 * MultiInstanceToPropositional.java
 * Copyright (C) 2005 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.filters.unsupervised.attribute;

import weka.core.Attribute;
import weka.core.Capabilities;
import weka.core.Instance;
import weka.core.Instances;
import weka.core.MultiInstanceCapabilitiesHandler;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.RelationalLocator;
import weka.core.SelectedTag;
import weka.core.StringLocator;
import weka.core.Tag;
import weka.core.Utils;
import weka.core.Capabilities.Capability;
import weka.filters.Filter;
import weka.filters.UnsupervisedFilter;

import java.util.Enumeration;
import java.util.Vector;

/** 
 <!-- globalinfo-start -->
 * Converts the multi-instance dataset into single instance dataset so that the Nominalize, Standardize and other type of filters or transformation  can be applied to these data for the further preprocessing.<br/>
 * Note: the first attribute of the converted dataset is a nominal attribute and refers to the bagId.
 * <p/>
 <!-- globalinfo-end -->
 * 
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -A &lt;num&gt;
 *  The type of weight setting for each prop. instance:
 *  0.weight = original single bag weight /Total number of
 *  prop. instance in the corresponding bag;
 *  1.weight = 1.0;
 *  2.weight = 1.0/Total number of prop. instance in the 
 *   corresponding bag; 
 *  3. weight = Total number of prop. instance / (Total number 
 *   of bags * Total number of prop. instance in the 
 *   corresponding bag). 
 *  (default:0)</pre>
 * 
 <!-- options-end -->
 *
 * @author Lin Dong ([e-mail]) 
 * @version $Revision: 1.6 $
 * @see PropositionalToMultiInstance
 */
public class MultiInstanceToPropositional 
  extends Filter 
  implements OptionHandler, UnsupervisedFilter, MultiInstanceCapabilitiesHandler {

  /** for serialization */
  private static final long serialVersionUID = -4102847628883002530L;

  /** the total number of bags */
  protected int m_NumBags;

  /** Indices of string attributes in the bag */
  protected StringLocator m_BagStringAtts = null;

  /** Indices of relational attributes in the bag */
  protected RelationalLocator m_BagRelAtts = null;
  
  /** the total number of the propositional instance in the dataset */
  protected int m_NumInstances;
  
  /** weight method: keep the weight to be the same as the original value */
  public static final int WEIGHTMETHOD_ORIGINAL = 0;
  /** weight method: 1.0 */
  public static final int WEIGHTMETHOD_1 = 1;
  /** weight method: 1.0 / Total # of prop. instance in the corresp. bag */
  public static final int WEIGHTMETHOD_INVERSE1 = 2;
  /** weight method: Total # of prop. instance / (Total # of bags * Total # of prop. instance in the corresp. bag) */
  public static final int WEIGHTMETHOD_INVERSE2 = 3;
  /** weight methods */
  public static final Tag[] TAGS_WEIGHTMETHOD = {
    new Tag(WEIGHTMETHOD_ORIGINAL, 
        "keep the weight to be the same as the original value"),
    new Tag(WEIGHTMETHOD_1, 
        "1.0"),
    new Tag(WEIGHTMETHOD_INVERSE1, 
        "1.0 / Total # of prop. instance in the corresp. bag"),
    new Tag(WEIGHTMETHOD_INVERSE2, 
        "Total # of prop. instance / (Total # of bags * Total # of prop. instance in the corresp. bag)")
  };

  /** the propositional instance weight setting method */
  protected int m_WeightMethod = WEIGHTMETHOD_INVERSE2;

  /**
   * Returns an enumeration describing the available options
   *
   * @return an enumeration of all the available options
   */
  public Enumeration listOptions() {
    Vector result = new Vector();
  
    result.addElement(new Option(
          "\tThe type of weight setting for each prop. instance:\n"
          + "\t0.weight = original single bag weight /Total number of\n"
          + "\tprop. instance in the corresponding bag;\n"
          + "\t1.weight = 1.0;\n"
          + "\t2.weight = 1.0/Total number of prop. instance in the \n"
          + "\t\tcorresponding bag; \n"
          + "\t3. weight = Total number of prop. instance / (Total number \n"
          + "\t\tof bags * Total number of prop. instance in the \n"
          + "\t\tcorresponding bag). \n"
          + "\t(default:0)",
          "A", 1, "-A <num>"));
    
    return result.elements();
  }


  /**
   * Parses a given list of options. <p/>
   * 
   <!-- options-start -->
   * Valid options are: <p/>
   * 
   * <pre> -A &lt;num&gt;
   *  The type of weight setting for each prop. instance:
   *  0.weight = original single bag weight /Total number of
   *  prop. instance in the corresponding bag;
   *  1.weight = 1.0;
   *  2.weight = 1.0/Total number of prop. instance in the 
   *   corresponding bag; 
   *  3. weight = Total number of prop. instance / (Total number 
   *   of bags * Total number of prop. instance in the 
   *   corresponding bag). 
   *  (default:0)</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 weightString = Utils.getOption('A', options);
    if (weightString.length() != 0) {
      setWeightMethod(
          new SelectedTag(Integer.parseInt(weightString), TAGS_WEIGHTMETHOD));
    } else {
      setWeightMethod(
          new SelectedTag(WEIGHTMETHOD_INVERSE2, TAGS_WEIGHTMETHOD));
    }	
  }

  /**
   * Gets the current settings of the classifier.
   *
   * @return an array of strings suitable for passing to setOptions
   */
  public String [] getOptions() {
    Vector        result;
    
    result = new Vector();
    
    result.add("-A");
    result.add("" + m_WeightMethod);

    return (String[]) result.toArray(new String[result.size()]);
  }

  /**
   * Returns the tip text for this property
   *
   * @return tip text for this property suitable for
   * displaying in the explorer/experimenter gui
   */
  public String weightMethodTipText() {
    return "The method used for weighting the instances.";
  }

  /**
   * The new method for weighting the instances.
   *
   * @param method      the new method
   */
  public void setWeightMethod(SelectedTag method){
    if (method.getTags() == TAGS_WEIGHTMETHOD)
      m_WeightMethod = method.getSelectedTag().getID();
  }

  /**
   * Returns the current weighting method for instances.
   * 
   * @return		the current weight method
   */
  public SelectedTag getWeightMethod(){
    return new SelectedTag(m_WeightMethod, TAGS_WEIGHTMETHOD);
  }

  /**
   * Returns a string describing this filter
   *
   * @return a description of the filter suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {

    return 
        "Converts the multi-instance dataset into single instance dataset "
      + "so that the Nominalize, Standardize and other type of filters or transformation "
      + " can be applied to these data for the further preprocessing.\n"
      + "Note: the first attribute of the converted dataset is a nominal "
      + "attribute and refers to the bagId.";
  }

  /** 
   * Returns the Capabilities of this filter.
   *
   * @return            the capabilities of this object
   * @see               Capabilities
   */
  public Capabilities getCapabilities() {
    Capabilities result = super.getCapabilities();

    // attributes
    result.disableAllAttributes();
    result.enable(Capability.NOMINAL_ATTRIBUTES);
    result.enable(Capability.RELATIONAL_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);
    
    // class
    result.enableAllClasses();
    result.enable(Capability.MISSING_CLASS_VALUES);
    
    // other
    result.enable(Capability.ONLY_MULTIINSTANCE);
    
    return result;
  }

  /**
   * Returns the capabilities of this multi-instance filter for the
   * relational data (i.e., the bags).
   *
   * @return            the capabilities of this object
   * @see               Capabilities
   */
  public Capabilities getMultiInstanceCapabilities() {
    Capabilities result = new Capabilities(this);

    // attributes
    result.enableAllAttributes();
    result.disable(Capability.RELATIONAL_ATTRIBUTES);
    result.enable(Capability.MISSING_VALUES);
    
    // class
    result.enableAllClasses();
    result.enable(Capability.MISSING_CLASS_VALUES);
    result.enable(Capability.NO_CLASS);
    
    // other
    result.setMinimumNumberInstances(0);
    
    return result;
  }

  /**
   * Sets the format of the input instances.
   *
   * @param instanceInfo an Instances object containing the input 
   * instance structure (any instances contained in the object are 
   * ignored - only the structure is required).
   * @return true if the outputFormat may be collected immediately
   * @throws Exception if the input format can't be set 
   * successfully
   */
  public boolean setInputFormat(Instances instanceInfo) 
    throws Exception {

    if (instanceInfo.attribute(1).type()!=Attribute.RELATIONAL) {
      throw new Exception("Can only handle relational-valued attribute!");
    }  
    super.setInputFormat(instanceInfo);   

    m_NumBags = instanceInfo.numInstances();
    m_NumInstances = 0;
    for (int i=0; i<m_NumBags; i++)
      m_NumInstances += instanceInfo.instance(i).relationalValue(1).numInstances();

    Attribute classAttribute = (Attribute) instanceInfo.classAttribute().copy();
    Attribute bagIndex = (Attribute) instanceInfo.attribute(0).copy();

    /* create a new output format (propositional instance format) */
    Instances newData = instanceInfo.attribute(1).relation().stringFreeStructure();
    newData.insertAttributeAt(bagIndex, 0);
    newData.insertAttributeAt(classAttribute, newData.numAttributes());
    newData.setClassIndex(newData.numAttributes() - 1);

    super.setOutputFormat(newData.stringFreeStructure());

    m_BagStringAtts = new StringLocator(instanceInfo.attribute(1).relation().stringFreeStructure());
    m_BagRelAtts    = new RelationalLocator(instanceInfo.attribute(1).relation().stringFreeStructure());

    return true;
  }


  /**
   * Input an instance for filtering. Filter requires all
   * training 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 set.
   */
  public boolean input(Instance instance) {

    if (getInputFormat() == null) {
      throw new IllegalStateException("No input instance format defined");
    }
    if (m_NewBatch) {
      resetQueue();
      m_NewBatch = false;
    }

    convertInstance(instance);
    return true;

  }

  /**
   * 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
   */
  public boolean batchFinished() {

    if (getInputFormat() == null) {
      throw new IllegalStateException("No input instance format defined");
    }

    Instances input = getInputFormat();

    // Convert pending input instances
    for(int i = 0; i < input.numInstances(); i++) {
      convertInstance(input.instance(i));
    }

    // Free memory
    flushInput();

    m_NewBatch = true;
    return (numPendingOutput() != 0);
  }

  /**
   * Convert a single bag over. The converted instances is 
   * added to the end of the output queue.
   *
   * @param bag the bag to convert
   */
  private void convertInstance(Instance bag) {

    Instances data = bag.relationalValue(1);
    int bagSize = data.numInstances();
    double bagIndex = bag.value(0);
    double classValue = bag.classValue();
    double weight = 0.0; 
    //the proper weight for each instance in a bag 
    if (m_WeightMethod == WEIGHTMETHOD_1)
      weight = 1.0;
    else if (m_WeightMethod == WEIGHTMETHOD_INVERSE1)
      weight = (double) 1.0 / bagSize;
    else if (m_WeightMethod == WEIGHTMETHOD_INVERSE2)
      weight=(double) m_NumInstances / (m_NumBags * bagSize);
    else 
      weight = (double) bag.weight() / bagSize;

    Instance newInst;
    Instances outputFormat = getOutputFormat().stringFreeStructure();

    for (int i = 0; i < bagSize; i++) {
      newInst = new Instance (outputFormat.numAttributes());
      newInst.setDataset(outputFormat);
      newInst.setValue(0,bagIndex);
      if (!bag.classIsMissing())
        newInst.setClassValue(classValue);
      // copy the attribute values to new instance
      for (int j = 1; j < outputFormat.numAttributes() - 1; j++){
        newInst.setValue(j,data.instance(i).value(j - 1));
      }	

      newInst.setWeight(weight);

      // copy strings/relational values
      StringLocator.copyStringValues(
	  newInst, false, 
	  data, m_BagStringAtts, 
	  outputFormat, m_OutputStringAtts);

      RelationalLocator.copyRelationalValues(
	  newInst, false, 
	  data, m_BagRelAtts, 
	  outputFormat, m_OutputRelAtts);
      
      push(newInst);
    }
  }

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