• Home
• nz.ac.waikato.cms.weka
• weka-dev
• 3.9.5
• source code
• FilteredClusterer.java

×

Project download

Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $

You can buy this project and download/modify it how often you want.

weka.clusterers.FilteredClusterer Maven / Gradle / Ivy

/*
 *   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 3 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, see .
 */

/*
 * FilteredClusterer.java
 * Copyright (C) 2006-2012 University of Waikato, Hamilton, New Zealand
 *
 */

package weka.clusterers;

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

import weka.core.*;
import weka.core.Capabilities.Capability;
import weka.filters.Filter;
import weka.filters.SupervisedFilter;

/**
 *  Class for running an arbitrary clusterer on data
 * that has been passed through an arbitrary filter. Like the clusterer, the
 * structure of the filter is based exclusively on the training data and test
 * instances will be processed by the filter without changing their structure.
 * 

 * 
 * 
 *  Valid options are:
 * 

 * 
 * 
 * -F <filter specification>
 *  Full class name of filter to use, followed
 *  by filter options.
 *  eg: "weka.filters.unsupervised.attribute.Remove -V -R 1,2"
 * (default: weka.filters.AllFilter)
 * 
 * 
 * 
 * -W
 *  Full name of base clusterer.
 *  (default: weka.clusterers.SimpleKMeans)
 * 
 * 
 * 
 * Options specific to clusterer weka.clusterers.SimpleKMeans:
 * 
 * 
 * 
 * -N <num>
 *  number of clusters.
 *  (default 2).
 * 
 * 
 * 
 * -V
 *  Display std. deviations for centroids.
 * 
 * 
 * 
 * -M
 *  Replace missing values with mean/mode.
 * 
 * 
 * 
 * -S <num>
 *  Random number seed.
 *  (default 10)
 * 
 * 
 * 
 * 
 * Based on code from the FilteredClassifier by Len Trigg.
 * 
 * @author Len Trigg ([email protected])
 * @author FracPete (fracpete at waikato dot ac dot nz)
 * @version $Revision: 15519 $
 * @see weka.classifiers.meta.FilteredClassifier
 */
public class FilteredClusterer extends SingleClustererEnhancer implements Drawable {

  /** for serialization. */
  private static final long serialVersionUID = 1420005943163412943L;

  /** The filter. */
  protected Filter m_Filter;

  /** The instance structure of the filtered instances. */
  protected Instances m_FilteredInstances;

  /**
   * Default constructor.
   */
  public FilteredClusterer() {
    m_Clusterer = new SimpleKMeans();
    m_Filter = new weka.filters.AllFilter();
  }

  /**
   * Returns a string describing this clusterer.
   * 
   * @return a description of the clusterer suitable for displaying in the
   *         explorer/experimenter gui
   */
  public String globalInfo() {
    return "Class for running an arbitrary clusterer on data that has been passed "
      + "through an arbitrary filter. Like the clusterer, the structure of the filter "
      + "is based exclusively on the training data and test instances will be processed "
      + "by the filter without changing their structure.";
  }

  /**
   * String describing default filter.
   * 
   * @return the default filter classname
   */
  protected String defaultFilterString() {
    return weka.filters.AllFilter.class.getName();
  }

  /**
   * Returns an enumeration describing the available options.
   * 
   * @return an enumeration of all the available options.
   */
  @Override
  public Enumeration listOptions() {
    Vector result = new Vector();

    result.addElement(new Option(
      "\tFull class name of filter to use, followed\n"
        + "\tby filter options.\n"
        + "\teg: \"weka.filters.unsupervised.attribute.Remove -V -R 1,2\"\n"
        + "(default: " + defaultFilterString() + ")", "F", 1,
      "-F "));

    result.addAll(Collections.list(super.listOptions()));

    return result.elements();
  }

  /**
   * Parses a given list of options.
   * 

   * 
   *  Valid options are:
   * 

   * 
   * 
   * -F <filter specification>
   *  Full class name of filter to use, followed
   *  by filter options.
   *  eg: "weka.filters.unsupervised.attribute.Remove -V -R 1,2"
   * (default: weka.filters.AllFilter)
   * 
   * 
   * 
   * -W
   *  Full name of base clusterer.
   *  (default: weka.clusterers.SimpleKMeans)
   * 
   * 
   * 
   * Options specific to clusterer weka.clusterers.SimpleKMeans:
   * 
   * 
   * 
   * -N <num>
   *  number of clusters.
   *  (default 2).
   * 
   * 
   * 
   * -V
   *  Display std. deviations for centroids.
   * 
   * 
   * 
   * -M
   *  Replace missing values with mean/mode.
   * 
   * 
   * 
   * -S <num>
   *  Random number seed.
   *  (default 10)
   * 
   * 
   * 
   * 
   * @param options the list of options as an array of strings
   * @throws Exception if an option is not supported
   */
  @Override
  public void setOptions(String[] options) throws Exception {
    String tmpStr;
    String[] tmpOptions;

    tmpStr = Utils.getOption('F', options);
    if (tmpStr.length() > 0) {
      tmpOptions = Utils.splitOptions(tmpStr);
      if (tmpOptions.length == 0) {
        throw new IllegalArgumentException(
          "Invalid filter specification string");
      }
      tmpStr = tmpOptions[0];
      tmpOptions[0] = "";
      setFilter((Filter) Utils.forName(Filter.class, tmpStr, tmpOptions));
    } else {
      setFilter(new weka.filters.AllFilter());
    }

    super.setOptions(options);
  }

  /**
   * Gets the current settings of the clusterer.
   * 
   * @return an array of strings suitable for passing to setOptions
   */
  @Override
  public String[] getOptions() {
    Vector result = new Vector();

    result.addElement("-F");
    result.addElement(getFilterSpec());

    Collections.addAll(result, super.getOptions());

    return 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 filterTipText() {
    return "The filter to be used.";
  }

  /**
   * Sets the filter.
   * 
   * @param filter the filter with all options set.
   */
  public void setFilter(Filter filter) {
    m_Filter = filter;

    if (m_Filter instanceof SupervisedFilter) {
      System.out
        .println("WARNING: you are using a supervised filter, which will leak "
          + "information about the class attribute!");
    }
  }

  /**
   * Gets the filter used.
   * 
   * @return the filter
   */
  public Filter getFilter() {
    return m_Filter;
  }

  /**
   * Gets the filter specification string, which contains the class name of the
   * filter and any options to the filter.
   * 
   * @return the filter string.
   */
  protected String getFilterSpec() {
    String result;
    Filter filter;

    filter = getFilter();
    result = filter.getClass().getName();

    if (filter instanceof OptionHandler) {
      result += " " + Utils.joinOptions(((OptionHandler) filter).getOptions());
    }

    return result;
  }

  /**
   * Returns default capabilities of the clusterer.
   * 
   * @return the capabilities of this clusterer
   */
  @Override
  public Capabilities getCapabilities() {
    Capabilities result;

    if (getFilter() == null) {
      result = super.getCapabilities();
      result.disableAll();
      result.enable(Capability.NO_CLASS);
    } else {
      result = getFilter().getCapabilities();
    }

    // set dependencies
    for (Capability cap : Capability.values()) {
      result.enableDependency(cap);
    }

    return result;
  }

  /**
   * Build the clusterer on the filtered data.
   * 
   * @param data the training data
   * @throws Exception if the clusterer could not be built successfully
   */
  @Override
  public void buildClusterer(Instances data) throws Exception {
    if (m_Clusterer == null) {
      throw new Exception("No base clusterer has been set!");
    }

    // remove instances with missing class
    if (data.classIndex() > -1) {
      data = new Instances(data);
      data.deleteWithMissingClass();
    }

    m_Filter.setInputFormat(data); // filter capabilities are checked here
    data = Filter.useFilter(data, m_Filter);

    // can clusterer handle the data?
    getClusterer().getCapabilities().testWithFail(data);

    m_FilteredInstances = data.stringFreeStructure();
    m_Clusterer.buildClusterer(data);
  }

  /**
   * Classifies a given instance after filtering.
   * 
   * @param instance the instance to be classified
   * @return the class distribution for the given instance
   * @throws Exception if instance could not be classified successfully
   */
  @Override
  public double[] distributionForInstance(Instance instance) throws Exception {

    if (m_Filter.numPendingOutput() > 0) {
      throw new Exception("Filter output queue not empty!");
    }

    if (!m_Filter.input(instance)) {
      throw new Exception(
        "Filter didn't make the test instance immediately available!");
    }

    m_Filter.batchFinished();
    Instance newInstance = m_Filter.output();

    return m_Clusterer.distributionForInstance(newInstance);
  }

  /**
   * Output a representation of this clusterer.
   * 
   * @return a representation of this clusterer
   */
  @Override
  public String toString() {
    String result;

    if (m_FilteredInstances == null) {
      result = "FilteredClusterer: No model built yet.";
    } else {
      result = "FilteredClusterer using " + getClustererSpec()
        + " on data filtered through " + getFilterSpec()
        + "\n\nFiltered Header\n" + m_FilteredInstances.toString()
        + "\n\nClusterer Model\n" + m_Clusterer.toString();
    }

    return result;
  }

  /**
   * Returns the type of graph this clusterer represents.
   *
   * @return the graph type of this clusterer
   */
  public int graphType() {

    if (m_Clusterer instanceof Drawable)
      return ((Drawable) m_Clusterer).graphType();
    else
      return Drawable.NOT_DRAWABLE;
  }

  /**
   * Returns graph describing the clusterer (if possible).
   *
   * @return the graph of the clusterer in dotty format
   * @throws Exception if the clusterer cannot be graphed
   */
  public String graph() throws Exception {

    if (m_Clusterer instanceof Drawable)
      return ((Drawable) m_Clusterer).graph();
    else
      throw new Exception(
              "Clusterer: " + getClustererSpec() + " cannot be graphed");
  }

  /**
   * Returns the revision string.
   * 
   * @return the revision
   */
  @Override
  public String getRevision() {
    return RevisionUtils.extract("$Revision: 15519 $");
  }

  /**
   * Main method for testing this class.
   * 
   * @param args the commandline options, use "-h" for help
   */
  public static void main(String[] args) {
    runClusterer(new FilteredClusterer(), args);
  }
}