weka.classifiers.neural.lvq.model.CodebookVector Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of wekaclassalgos
Fork of the following defunct sourceforge.net project: https://sourceforge.net/projects/wekaclassalgos/
There is a newer version: 2023.2.8
/*
 *   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 .
 */

package weka.classifiers.neural.lvq.model;

import weka.core.Instance;
import weka.core.Utils;

import java.io.Serializable;

/**
 * Description: Represents a single codebook vector in an LVQ model
 * A codebook vector is also called a prototype or an exemplar. It is
 * a single node which represents a sign post in the state space
 * of the training data
 * 
 * 

 * Copyright (c) Jason Brownlee 2004
 * 
 *
 * @author Jason Brownlee
 */
public class CodebookVector
  implements Serializable {

  protected final int codebookId;

  /**
   * Vecotrs attribute values
   */
  protected double[] attributes;

  /**
   * Index of the class attribute
   */
  protected int classAttributeIndex;

  /**
   * Distance from a specific data instance at a moment in time
   */
  protected double distance;

  /**
   * Used in cases where codebook vectors can have their own individual learning rate
   */
  protected double learningRate;

  /**
   * The number of times that the codebook vector is the bmu
   */
  protected int bmuCorrectCount;

  protected int bmuIncorrectCount;

  private boolean voting;

  private int[] classHitDistribution;


  public CodebookVector(int aCodebookId) {
    codebookId = aCodebookId;
  }

  public void setClassification(double aClassificationValue) {
    attributes[classAttributeIndex] = aClassificationValue;
  }

  public void initialise(double[] aAttributes, int aClassIndex, int aNumClasses) {
    attributes = aAttributes;
    classAttributeIndex = aClassIndex;
    classHitDistribution = new int[aNumClasses];
  }


  /**
   * String representation of this codebook vector
   *
   * @return String
   */
  public String toString() {
    StringBuffer buffer = new StringBuffer(100);

    // all attributes
    for (int i = 0; i < attributes.length; i++) {
      buffer.append(attributes[i]);
      buffer.append(", ");
    }

    // class index
    buffer.append(getClassification());

    return buffer.toString();
  }

  /**
   * Return codebook vectors class assignmnet
   *
   * @return
   */
  public double getClassification() {
    if (voting) {
      int largestIndex = 0;

      // find index with largest value
      for (int i = 1; i < classHitDistribution.length; i++) {
	if (classHitDistribution[i] > classHitDistribution[largestIndex]) {
	  largestIndex = i;
	}
      }

      // check for all empty
      if (classHitDistribution[largestIndex] == 0) {
	return attributes[classAttributeIndex];
      }
      else {
	return (double) largestIndex;
      }
    }

    return attributes[classAttributeIndex];
  }

  /**
   * Get distance from data instance at a point in time
   *
   * @return
   */
  public double getDistance() {
    return distance;
  }

  /**
   * Return codebook vector's internal representation
   *
   * @return
   */
  public double[] getAttributes() {
    return attributes;
  }

  /**
   * codebook vectors individual learning rate
   *
   * @return
   */
  public double getIndividualLearningRate() {
    return learningRate;
  }

  /**
   * Set the codebook vectors learning rate
   *
   * @param lrate
   */
  public void setIndividualLearningRate(double lrate) {
    learningRate = lrate;
  }

  public double value(int aIndex) {
    return attributes[aIndex];
  }

  public void setBmuHit(double aDistance, Instance aInstance) {
    distance = aDistance;

    // check for training mode
    if (!Utils.isMissingValue(aInstance.classValue())) {
      if (aInstance.classValue() == getClassification()) {
	bmuCorrectCount++;
      }
      else {
	bmuIncorrectCount++;
      }

      // store for class distribution
      classHitDistribution[(int) aInstance.classValue()]++;
    }
  }

  public void resetBmuCounts() {
    bmuCorrectCount = bmuIncorrectCount = 0;
  }

  public int getBmuCorrectCount() {
    return bmuCorrectCount;
  }

  public int getBmuIncorrectCount() {
    return bmuIncorrectCount;
  }

  public int getId() {
    return codebookId;
  }

  public int[] getClassHitDistribution() {
    return classHitDistribution;
  }

  public boolean hasClassChanged() {
    return (getClassification() != attributes[classAttributeIndex]);
  }

  public void setUseVoting(boolean useVoting) {
    voting = useVoting;
  }

  public void clearClassDistributions() {
    for (int i = 0; i < classHitDistribution.length; i++) {
      classHitDistribution[i] = 0;
    }
  }
}