weights.java

利用Java实现的神经网络工具箱

/*
 * Weights.java	1.0 10 Jun 2004
 *
 * NeuralNetworkToolkit
 * Copyright (C) 2004 Universidade de Bras�lia
 *
 * This file is part of NeuralNetworkToolkit.
 *
 * NeuralNetworkToolkit 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.
 *
 * NeuralNetworkToolkit 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 NeuralNetworkToolkit; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA - 02111-1307 - USA.
 */

package neuralnetworktoolkit.neuralnetwork;

import java.io.Serializable;
import java.util.Vector;

/**
 * Class that implements the data structure that stores the neural
 * network weights information. <br><b>Dynamic and recurrent cases
 * stay unimplemented!<b>
 * 
 * @version 1.0 09 Jun 2004
 * 
 * @author <a href="mailto:[e-mail]">Hugo Iver V. Gon�alves</a>
 * @author <a href="mailto:[e-mail]">Rodrigo C. M. Coimbra</a>
 */
public class Weights implements Serializable {

	/** Indicates that weights list is dynamic or static. */
	private int      isDynamic;
	/** Indicates that weigths list is recurrent or not. */
	private int      isRecurrent;
	/** Weights list size. */
	private int      size                     = 0;
	/** Stores not recurrent weights list dynamically. */
	private Vector   dynamicWeigths;
	/** Stores recurrent weights list dynamically. */
	private Vector   dynamicRecurrentWeights;
	/** <i>Bias</i> term value. */
	private double   bias;
	/** Stores not recurrent weights list statically. */
	private double[] staticWeights;
	/** Stores recurrent weights list statically. */ 
	private double[] staticRecurrentWeights;

	/**
	 * Creates a new weights list with the specified attributes.
	 * 
	 * @param size      Weights list size.
	 * @param dynamic   Indicates that the weights list is dynamic or static.
	 * @param recurrent Indicates that the weights list is recurrent or not.
	 */
	public Weights(int size, int dynamic, int recurrent) {
		this.isDynamic = dynamic;
		this.isRecurrent = recurrent;
		this.size = size;
		//	Outer switch
		switch (isDynamic) {
			case NeuralNetwork.NOT_DYNAMIC :
				{
					// 1st Inner switch	
					switch (isDynamic) {
						case NeuralNetwork.NOT_RECURRENT :
							{
								staticWeights = new double[size];
							}
							break;
						case NeuralNetwork.RECURRENT :
							{
								// TODO implement this case.
							}
							break;
					} // end 1st Inner switch	
				}
				break;
			case NeuralNetwork.DYNAMIC :
				{ // TODO implement the dynamic case.
					// 2nd Inner switch	
					switch (isDynamic) {
						case NeuralNetwork.NOT_RECURRENT :
							{
								// TODO Implement this.
							}
							break;
						case NeuralNetwork.RECURRENT :
							{
								//	TODO implement this case.
							}
							break;
					} // end 2nd Inner switch
				}
				break;
		} // end Outer switch

	} //Weights()

	/**
	 * Returns the weight at index.
	 * 
	 * @param index Weight index.
	 * 
	 * @return Weight value.
	 */
	public double getWeight(int index) {
		// TODO Implement dynamic case.
		return staticWeights[index];
		
	} //getWeight()

	/**
	 * Sets the weight value at index.
	 * 
	 * @param index     Weight index to be seted.
	 * @param newWeight New weight value.
	 */
	public void setWeight(int index, double newWeight) {
		// TODO Implement dynamic case.
		staticWeights[index] = newWeight;
		
	} //setWeight()

	/**
	 * Updates the weight value at index.
	 * 
	 * @param index     Weight index to be updated.
	 * @param increment Value to be incremented at weight value.
	 */
	public void updateWeight(int index, double increment) {
		// TODO Implement dynamic case.
		staticWeights[index] = staticWeights[index] + increment;
		
	} //updateWeight()

	/**
	 * Sets initial weights with random values.
	 */
	/*public void randomicWeightStart() {
		switch (isDynamic) {
			case NeuralNetwork.NOT_DYNAMIC :
				{
					for (int i = 0; i < staticWeights.length; i++) {
						staticWeights[i] = Math.random() * 10;
					}
				}
				break;
			case NeuralNetwork.DYNAMIC :
				{
					// TODO Implement this case.
				}
				break;
		}

	} //randomicWeightStart()*/

	/**
	 * Updates <i>bias</i> term with an increment value.
	 * 
	 * @param increment Value to increment <i>bias</i>.
	 */
	public void updateBias(double increment){
		bias = bias + increment;
		
	} //updateBias()

	/**
	 * Sets the wished initial weights value.
	 * 
	 * @param weight Initial weights value to be seted.
	 */
	public void startWeights(int weight) {
		switch (isDynamic) {
			case NeuralNetwork.NOT_DYNAMIC :
				{
					for (int i = 0; i < staticWeights.length; i++)
						staticWeights[i] = weight;
					
				}
				break;
			case NeuralNetwork.DYNAMIC :
				{
					// TODO Implement this.
				}
				break;
		}

	} //startWeights()

	/**
	 * Sets initial values (randomically) to <i>bias</i> term and weights.
	 */
	/*public void randomicBiasAndWeightStart() {
		bias = Math.random() - 0.5;
		switch (isDynamic) {
			case NeuralNetwork.NOT_DYNAMIC :
				{
					for (int i = 0; i < staticWeights.length; i++)
						staticWeights[i] = (Math.random() - 0.5) * 10;
					
				}
				break;
			case NeuralNetwork.DYNAMIC :
				{

				}
				break;
		}

	} //randomicBiasAndWeightStart()*/

	/**
	 * Sets initial <i>bias</i> term with a random value.
	 */
	/*public void randomicStartBias() {
		bias = Math.random() * 10;
		
	} //randomicStartBias()*/
	
	
	/**
	 * Sets the wished initial <i>bias</i> term value.
	 * 
	 * @param bias Initial <i>bias</i> value to be seted.
	 */
	public void startBias(double bias) {
		this.bias = bias;
		
	} //startBias()

	/**
	 * Sets weights list to be dynamic or static.
	 * 
	 * @param condition Indicates that list is dynamic or static.
	 */
	public void setIsDynamic(int condition) {
		isDynamic = condition;
		
	} //setIsDynamic()

	/**
	 * Returns <i>bias</i> term value.
	 * 
	 * @return <i>Bias</i> term value.
	 */
	public double getBias() {
		return bias;
		
	} //getBias()

	/**
	 * Sets <i>bias</i> term value.
	 * 
	 * @param bias Value to be seted.
	 */
	public void setBias(double bias) {
		this.bias = bias;
		
	} //setBias()

	/**
	 * Returns weights list size.
	 * 
	 * @return Weights list size.
	 */
	public int getSize() {
		return size;
		
	} //getSize()

} //Weights